Wireless Security Handbook

AU3378_TitlePage 11/16/05 8:59 AM Page 1 Wireless Security Handbook Aaron E. Earle Boca Raton New York © 2006 by Tay...

Author: Aaron E. Earle

41 downloads 3175 Views 17MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

AU3378_TitlePage 11/16/05 8:59 AM Page 1

Wireless Security Handbook

Aaron E. Earle

Boca Raton New York

© 2006 by Taylor & Francis Group, LLC

AU3378_Discl.fm Page 1 Monday, September 26, 2005 3:54 PM

Published in 2006 by Auerbach Publications Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2006 by Taylor & Francis Group, LLC Auerbach is an imprint of Taylor & Francis Group No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number-10: 0-8493-3378-4 (Hardcover) International Standard Book Number-13: 978-0-8493-3378-1 (Hardcover) Library of Congress Card Number 2005049924 This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data Earle, Aaron E. Wireless security handbook / Aaron E. Earle. p. cm. Includes bibliographical references and index. ISBN 0-8493-3378-4 (alk. paper) 1. Wireless LANs--Security measures. 2. Wireless communication systems--Security measures. I. Title. TK5105.78.E23 2005 005.8--dc22

2005049924

Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com Taylor & Francis Group is the Academic Division of Informa plc.


and the Auerbach Publications Web site at http://www.auerbach-publications.com

OTHER INFORMATION SECURITY BOOKS FROM AUERBACH Asset Protection and Security Management Handbook POA Publishing ISBN: 0-8493-1603-0

Information Technology Control and Audit Fredrick Gallegos, Daniel Manson, and Sandra Allen-Senft ISBN: 0-8493-9994-7

Building a Global Information Assurance Program Raymond J. Curts and Douglas E. Campbell ISBN: 0-8493-1368-6

Investigator’s Guide to Steganography Gregory Kipper ISBN: 0-8493-2433-5

Building an Information Security Awareness Program Mark B. Desman ISBN: 0-8493-0116-5 Critical Incident Management Alan B. Sterneckert ISBN: 0-8493-0010-X Cyber Crime Investigator’s Field Guide Bruce Middleton ISBN: 0-8493-1192-6 Cyber Forensics: A Field Manual for Collecting, Examining, and Preserving Evidence of Computer Crimes Albert J. Marcella, Jr. and Robert S. Greenfield ISBN: 0-8493-0955-7 The Ethical Hack: A Framework for Business Value Penetration Testing James S. Tiller ISBN: 0-8493-1609-X The Hacker’s Handbook: The Strategy Behind Breaking into and Defending Networks Susan Young and Dave Aitel ISBN: 0-8493-0888-7 Information Security Architecture: An Integrated Approach to Security in the Organization Jan Killmeyer Tudor ISBN: 0-8493-9988-2 Information Security Fundamentals Thomas R. Peltier ISBN: 0-8493-1957-9 Information Security Management Handbook, 5th Edition Harold F. Tipton and Micki Krause ISBN: 0-8493-1997-8 Information Security Policies, Procedures, and Standards: Guidelines for Effective Information Security Management Thomas R. Peltier ISBN: 0-8493-1137-3 Information Security Risk Analysis, 2nd Edition Thomas R. Peltier ISBN: 0-8493-3346-6

Managing a Network Vulnerability Assessment Thomas Peltier, Justin Peltier, and John A. Blackley ISBN: 0-8493-1270-1 Network Perimeter Security: Building Defense In-Depth Cliff Riggs ISBN: 0-8493-1628-6 The Practical Guide to HIPAA Privacy and Security Compliance Kevin Beaver and Rebecca Herold ISBN: 0-8493-1953-6 A Practical Guide to Security Engineering and Information Assurance Debra S. Herrmann ISBN: 0-8493-1163-2 The Privacy Papers: Managing Technology, Consumer, Employee and Legislative Actions Rebecca Herold ISBN: 0-8493-1248-5 Public Key Infrastructure: Building Trusted Applications and Web Services John R. Vacca ISBN: 0-8493-0822-4 Securing and Controlling Cisco Routers Peter T. Davis ISBN: 0-8493-1290-6 Strategic Information Security John Wylder ISBN: 0-8493-2041-0 Surviving Security: How to Integrate People, Process, and Technology, Second Edition Amanda Andress ISBN: 0-8493-2042-9 A Technical Guide to IPSec Virtual Private Networks James S. Tiller ISBN: 0-8493-0876-3 Using the Common Criteria for IT Security Evaluation Debra S. Herrmann ISBN: 0-8493-1404-6

AUERBACH PUBLICATIONS www.auerbach-publications.com To Order Call: 1-800-272-7737 • Fax: 1-800-374-3401 E-mail: [email protected]


AU3378_C000.fm Page v Thursday, November 17, 2005 12:04 PM

Contents 1 Wireless Network Overview 1.1 1.2

1.3 1.4

1.5

1.6

RF Overview Wireless Signal Propagation 1.2.1 Reflection 1.2.2 Refraction 1.2.3 Diffraction 1.2.4 Scattering 1.2.5 Absorption Signal-to-Noise Ratio Modulation 1.4.1 Amplitude Modulation 1.4.2 Frequency Modulation 1.4.3 Phase Modulation 1.4.4 Complementary Code Keying (CCK) 1.4.5 Quadrature Amplitude Modulation (QAM) Wireless Groups 1.5.1 International Telecommunications Union (ITU) 1.5.2 International Telecommunications Union Radio Sector (ITU-R) 1.5.3 Federal Communications Commission (FCC) 1.5.4 Conference of European Post and Telecommunications (CEPT) 1.5.5 Wi-Fi Alliance 1.5.6 IEEE Chapter 1 Review Questions

2 Risks and Threats of Wireless 2.1

2.2 2.3

Goals of Information Security 2.1.1 Confidentiality 2.1.2 Availability 2.1.3 Integrity Analysis Spoofing

v © 2006 by Taylor & Francis Group, LLC

AU3378_C000.fm Page vi Thursday, November 17, 2005 12:04 PM

vi


2.4 2.5 2.6 2.7 2.8 2.9

2.10 2.11

Denial-of-Service Malicious Code.. Social Engineering Rogue Access Points Cell Phone Security Wireless Hacking and Hackers 2.9.1 Motives of Wireless Hackers 2.9.2 War Drivers 2.9.3 War Walkers 2.9.4 War Chalking 2.9.5 War Flying 2.9.6 Bluejacking 2.9.7 X10 Driving 2.9.8 Cordless Phone Driving 2.9.9 War Dialing 2.9.10 Tracking War Drivers RFID Chapter 2 Review Questions

3 The Legality of Computer Crime 3.1 3.2 3.3 3.4

Electronic Communications Privacy Act Computer Fraud and Abuse Act 3.2.1 Patriot Act State Computer Crime Issues Chapter 3 Review Questions

4 Wireless Physical Layer Technologies 4.1 4.2 4.3 4.4 4.5

ISM Spectrum Frequency Hopping Spread Spectrum (FHSS) Direct Sequence Spread Spectrum (DSSS) Orthogonal Frequency Division Multiplexing (OFDM) Chapter 4 Review Questions

5 Wireless Management Frames 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8

Beacon Probe Request Probe Response Authentication Association Request Association Response Disassociation and De-Authentication CSMA/CA 5.8.1 RTS 5.8.2 CTS 5.8.3 DATA 5.8.4 ACK


AU3378_C000.fm Page vii Thursday, November 17, 2005 12:04 PM

Contents 5.9 5.10 5.11 5.12 5.13 5.14

Fragmentation Distributed Coordination Function Point Coordination Function Interframe Spacing Service Set Identifier (SSID) Chapter 5 Review Questions

6 Wireless Local and Personal Area Networks 6.1 6.2 6.3 6.4 6.5 6.6

6.7

6.8

Ad Hoc Mode Infrastructure Mode Bridging Repeater Mesh Wireless Networks Local Area Networking Standards 6.6.1 802.11 6.6.2 802.11a. 6.6.3 802.11b 6.6.4 802.11c 6.6.5 802.11d 6.6.6 802.11e 6.6.7 802.11f 6.6.8 802.11g 6.6.9 802.11h 6.6.10 802.11i 6.6.11 802.11j 6.6.12 802.11n 6.6.13 Real-World Wireless Data Rates Personal Area Network (PAN) 802.15 6.7.1 Bluetooth 802.15.1 6.7.2 Infrared (IR) 6.7.3 Ultrawide Band 802.15.3 6.7.4 ZIGBEE 802.15.4 Chapter 6 Review Questions

7 Wide Area Wireless Technologies 7.1

7.2 7.3

Cell Phone Technologies 7.1.1 Analog 7.1.2 TDMA 7.1.3 CDMA 7.1.3.1 CDMA2000 7.1.3.2 CDMA 1xEV-DO and CDMA 1xEV-DV 7.1.4 GSM. 7.1.4.1 GPRS. 7.1.4.2 GSM Security System Overview GPS 802.16 Air Interface Standard


vii

AU3378_C000.fm Page viii Thursday, November 17, 2005 12:04 PM

viii Wireless Security Handbook 7.4 7.5

802.20 Standard Chapter 7 Review Questions .

8 Wireless Antenna Theory 8.1

8.2 8.3

8.4

8.5

RF Antenna Overview 8.1.1 Polarization 8.1.2 Gain 8.1.2.1 Equivalent Isotropic Radiated Power (EIRP). 8.1.3 Beamwidth 8.1.4 Path Loss 8.1.5 Azimuth 8.1.6 Multipath 8.1.7 Antenna Diversity Fresnel Zone Antenna Types 8.3.1 Directional Antennas. 8.3.2 Omni-Directional Antennas 8.3.3 Homemade Antennas Connectors 8.4.1 N Connectors. 8.4.2 Reverse-Polarity TNC-Type Connector (RP-TNC) 8.4.3 SMA, RP-SMA, and RSMA 8.4.4 MC and MMX Chapter 8 Review Questions

9 The Wireless Deployment Process 9.1 9.2 9.3 9.4

9.5 9.6 9.7 9.8 9.9 9.10

Gather Requirements Estimation. Make the Business Case Site Survey 9.4.1 Performing the Site Survey 9.4.2 Technical Controls 9.4.3 Financial Controls . Design Staging Deployment and Installation. Certification Audit Chapter 9 Review Questions

10 Wireless Access Points 10.1 10.2

10.3

Linksys Access Points Cisco Access Points 10.2.1 Cisco Aironet 350 Series. 10.2.2 Cisco 1200 Series Access Point 10.2.3 Cisco 1100 Series Access Point Chapter 10 Review Questions


AU3378_C000.fm Page ix Thursday, November 17, 2005 12:04 PM

Contents

11 Wireless End Devices 11.1 11.2 11.3

11.4 11.5 11.6 11.7

Laptops Tablets PDA Devices 11.3.1 Palm 11.3.2 Microsoft CE and Pocket PC 11.3.3 BlackBerry RIM OS 11.3.4 Symbian OS 11.3.5 Linux Handheld Scanners Smart Phones Wi-Fi Phones Chapter 11 Review Questions

12 Wireless LAN Security 12.1 12.2

Wireless LAN Security History Authentication 12.2.1 Shared Key Authentication 12.2.2 Open Key Authentication 12.3 SSID 12.4 Wireless Security Basics 12.5 Equivalent Privacy Standard (WEP) 12.5.1 WEP Encryption Process 12.6 802.1x 12.6.1 Authentication Server 12.6.2 Authenticator 12.6.3 Supplicant 12.6.4 Extensive Authentication Protocol over Local Area Network (EAPOL) 12.7 Remote Authentication Dial-In User Service (RADIUS) 12.8 Extensible Authentication Protocol (EAP) 12.8.1 EAP-MD5 12.8.2 EAP-TLS 12.8.3 EAP-TTLS 12.8.4 LEAP 12.8.5 PEAP 12.8.6 EAP-FAST 12.9 Wi-Fi Protected Access (WPA) 12.10 802.11i 12.10.1 Robust Secure Network (RSN) 12.10.1.1 Transition Secure Network (TSN) 12.10.2 Temporal Key Integrity Protocol (TKIP) 12.10.2.1 TKIP Message Integrity Check (MIC) 12.10.3 Advanced Encryption Standard (AES) 12.10.4 802.11i System Overview 12.11 Wi-Fi Protected Access (WPA2) 12.12 WLAN Authentication and Privacy Infrastructure (WAPI) © 2006 by Taylor & Francis Group, LLC

ix

AU3378_C000.fm Page x Thursday, November 17, 2005 12:04 PM

x


12.13 Rogue Access Points Detection 12.14 Chapter 12 Review Questions

13 Breaking Wireless Security 13.1

13.2

13.3

13.4

The Hacking Process 13.1.1 Information Gathering 13.1.2 Enumeratio 13.1.3 Compromise 13.1.4 Expanding Privileges and Accessibility 13.1.5 Cleaning up the Trails Wireless Network Compromising Techniques 13.2.1 WEP 13.2.1.1 Stream Cipher Attack 13.2.1.2 Known Plaintext Attack 13.2.1.3 Dictionary Building Attack 13.2.1.4 Double Encryption Attack 13.2.1.5 Message Modification Attack 13.2.2 Denial-of-Service (DoS) Attacks 13.2.2.1 EAP DoS Attacks 13.2.3 MAC Filtering Attack 13.2.4 Cisco LEAP Vulnerabilities 13.2.5 RADIUS Vulnerabilities 13.2.6 802.1x Vulnerabilities 13.2.7 Attack on Michael 13.2.8 Attacks on Wireless Gateways 13.2.9 Attacks on WPA and 802.11i Access Point Compromising Techniques 13.3.1 Remote Management Attacks 13.3.1.1 Telnet 13.3.1.2 HTTP 13.3.1.3 RADIUS 13.3.1.4 SNMP Chapter 13 Review Questions

14 Wireless Security Policy 14.1

14.2 14.3

14.4

Policy Overview 14.1.1 Policies 14.1.2 Standards 14.1.3 Guidelines 14.1.4 Procedures The Policy-Writing Process Risk Assessment 14.3.1 Exposure Factor (EF) 14.3.2 Annualized Rate of Occurrence (ARO) 14.3.4 Single Loss Expectancy (SLE) 14.3.5 Annualized Loss Expectancy (ALE) Impact Analysis


AU3378_C000.fm Page xi Thursday, November 17, 2005 12:04 PM

Contents 14.5

14.6

Wireless Security Policy Areas 14.5.1 Password Policy 14.5.2 Access Policy 14.5.3 Public Access 14.5.4 Physical Security Chapter 14 Review Questions

15 Wireless Security Architectures 15.1 15.2

15.3 15.4 15.5 15.6 15.7

15.8

Static WEP Wireless Architecture VPN 15.2.1 Technology Overview. 15.2.1.1 IPSec 15.2.1.2 ISAKMP 15.2.1.3 Internet Key Exchange (IKE) 15.2.1.4 AH 15.2.1.5 ESP Wireless VPN Architecture Overview VPN Policy Aspect Wireless Gateway Systems 802.1x Comparing Wireless Security Architectures 15.7.1 WEP Architecture 15.7.2 Wireless VPN Architecture 15.7.3 Wireless Gateway or Firewall Architecture 15.7.4 Wireless 802.1x Architecture Chapter 15 Review Questions

16 Wireless Tools 16.1

16.2

16.3

16.4

16.5

Scanning Tools. 16.1.1 Network Stumbler 16.1.2 MiniStumbler 16.1.3 Wellenreiter 16.1.4 Wavemon Sniffing Tools 16.2.1 AiroPeek 16.2.2 Sniffer Pro 16.2.3 Mognet Hybrid Tools 16.3.1 Kismet 16.3.2 AirTraf 16.3.3 AirMagnet Denial-of-Service Tools 16.4.1 WLAN-Jack 16.4.2 FATA-Jack. Cracking Tools . 16.5.1 WEPCrack 16.5.2 AirSnort


xi

AU3378_C000.fm Page xii Thursday, November 17, 2005 12:04 PM

xii

16.6

16.7

16.8

Wireless Security Handbook 16.5.3 16.5.4 Access 16.6.1 16.6.2

BSD-Airtools ASLEAP Point Attacking Tools Brutus SolarWinds 16.6.2.1 Port Scanner Tool. 16.6.2.2 SNMP Brute Force Attack Tool. 16.6.2.3 SNMP Dictionary Attack Tool. 16.6.2.4 Router Password Decryption Tool 16.6.3 Cain and Able Other Wireless Security Tools 16.7.1 EtherChange 16.7.2 Achilles Chapter 16 Review Questions

Appendix A: Review Question Answers


AU3378_C000.fm Page xiii Thursday, November 17, 2005 12:04 PM

Preface This book was written to give the reader a well-rounded understanding of wireless network security. It looks at wireless from multiple perspectives, ranging from auditor, to security architect, to hacker. This wide scope benefits anyone who has to administer, secure, hack, or participate on a wireless network. Going through this book, the reader will see that it tackles the risk of wireless from many angles. It goes from a policy level to mitigate certain risks that wireless brings. It talks about the most cost-effective solutions to deploy wireless across a large enterprise. It talks about financial and technical controls that one can apply to reduce any unforeseen risk involved in a large wireless project. It covers the technical details of how to design, build, and hack almost all wireless security methods. The wide scope of knowledge that this book brings will help one become acquainted with the many aspects of wireless communications. This book also has career advancement in mind by covering all the objectives of the three widely upheld wireless certifications currently on the market. These certifications are administered by Planet3 Wireless and Cisco Systems. The focus of this book is on wireless local area networking technologies to meet these objectives, although this book looks at the security of almost all mobile communications. So if you are interested in obtaining a certification or just a deep knowledge of wireless security this book is for you.

xiii © 2006 by Taylor & Francis Group, LLC

AU3378_C000.fm Page xv Thursday, November 17, 2005 12:04 PM

Acknowledgments I would like to thank many people who over the years have helped me get to where I am today. Great wisdom comes from one who knows that it is not what you do to advance, but rather what the people below you do to push you in that direction. I would like to thank my family and friends who have supported me throughout this endeavor, and my girlfriend Clare who did not complain about the long hours away from her spent writing this book. I would like to thank my father Douglas R. Earle, who purchased my first computer for me; my friend Justin Peltier, who gave me the “I can do it, you can do it” mentality; and my friend Paul Immo, who saw my passion for technology and helped me achieve my goals around education and certification. I would also like to thank my friend Jeremy Davison for allowing me to forget altogether about computers, networking, security, and technology and just have fun every now and then.

xv © 2006 by Taylor & Francis Group, LLC

AU3378_book.fm Page 1 Monday, November 7, 2005 6:51 PM

Chapter 1

Wireless Network Overview This chapter looks at radio frequencies (RF) in general. The goal of this chapter is to gain a general understanding of RF. This allows us to see what issues are inherent in all wireless communications, whether it is a cell phone or an 802.11g laptop. This knowledge can help us troubleshoot RF networks and understand what can and cannot be fixed. After reading this, we look at the many types of interference that affects all wireless communications. Once an understanding of interference is achieved, we look at modulation. We discuss the different types of modulation used on wireless networks and how each of them works. The final section of this chapter addresses the many wireless groups that create and regulate the way we use wireless communications.

1.1 RF Overview What are radio frequencies, and where did they come from? Radio frequencies are nothing more than power, in the form of an alternating current created by an electrical device that passes though wiring and out an antenna. The antenna then radiates this power, creating radio waves that travel across the air in all directions until the waves become so minute that one cannot detect them. Heinrich Hertz discovered radio transmission

1 © 2006 by Taylor & Francis Group, LLC


2


in the late 1880s; he expounded on James Clerk Maxwell’s research on the electromagnetic theory of light. Hertz found that by using a strong electrical signal it was possible to send that signal through nonconductive material; later, the notion of such material went out the door when Hertz discovered that the signal could conduct through the air. This is how radio signals and thus wireless communications were born. As the radio waves travel across the air, a receiving antenna can take the wave and convert it back to an electrical signal. This signal would be the same as the one originally created by the sending electrical device. The way wireless propagates itself is very similar to dropping a stone into a large body of water. Once the stone hits the water, ripples are created, moving in all directions until the ripples are so minuscule that they no longer can be seen or detected. Electromagnetic waves are produced by the motion of electrically charged particles. These waves are also called electromagnetic radiation because they radiate from the electrically charged particles. All wireless devices have some form of electromagnetic waves. All these waves are part of the electromagnetic spectrum; this spectrum has all types of electromagnetic radiation classified. Although the size of this spectrum is infinite, the size of the radio portion is limited to around 100 kHz to 300 GHz. The waves discussed herein are mostly based in the microwave section of the radio spectrum. The larger an electromagnetic wave, the further it will travel. The fact is that when you look at radio waves, the amount of information being sent is small, and therefore the frequency used is also small. A small frequency signal has a very large wave. A radio wave, like the ones one picks up on a car radio, can be thought of as about the size of an adult elephant. Now look at an x-ray wave. This is very high on the radio spectrum, so it will have a large amount of data traveling down a small wavelength. This wave might be as small or smaller than a single atom. This smaller x-ray wave will not travel as far as the radio wave because of its limited size. In discussing frequency, one must understand how to measure it. When looking at a wave traveling in time, one can see the amount of times a signal wave is completed from an upper crest to its lower crest. Each time this is completed, it is a single cycle. When one measures the total amount of wave cycles in a particular amount of time, one gets a frequency. In general, one takes the amount of cycles in a single second, giving the hertz (Hz). In the case of wireless networks, this amount is so large that it is measured in gigahertz (GHz), which is one billion hertz. When talking about power and wireless, there are a number of values commonly used to measure wireless power. The first value to look at is the Watt, the rate at which a device converts electrical energy into another



Wireless Network Overview

3

form of energy, such as light, heat, or — in this case — a wireless signal. The Watt can be measured in a number of ways, depending on how high or low a value it is compared to a single watt. What this means is if one has a value much greater than a single watt, maybe somewhere around 1000 watts, one would have a kilowatt (kW). This is because a kilowatt represents 1000 watts. Now, if one has less than a single watt, then one has a milliwatt (mW), which is 1/1000 of a watt. The milliwatt is the primary watt designation in relation to wireless local area networking. The next term is the decibel. A decibel (or dB) is a mathematical — or, to be specific, a logarithmic ratio — that indicates the relative strength of a device’s electric or acoustic signal to that of another. This can be used by itself, although it is mostly used with a unit of measurement. Looking at wireless, the most common units of measurement used with the decibel are the milliwatt (dBm), the forward gain of an antenna compared to an imaginary isotropic antenna (dBi), and the forward gain of an antenna compared to a half-wave dipole antenna (dBd). Wireless networks are measured in decibel strength compared to one milliwatt. In wireless local area networking (WLAN), dBi and dBd are commonly used and a formula is often needed to convert these two expressions into each other so they an be correctly compared. Chapter 8 goes into greater detail about both isotropic and dipole antennas and power measurement. Until then, just remember that these two figures are the most commonly used measurements of wireless power. When discussing bandwidth, most computer people associate it with network performance. In the wireless world, bandwidth has a slightly different meaning. The meaning we are looking for in relation to wireless has to do with the size or the upper and lower limit to the frequency we are using. When we compare frequency and bandwidth, we see that frequency is a specific location on the electromagnetic spectrum compared to wireless bandwidth, which is the range between two frequencies. A single channel on the 2.4-GHz frequency has a channel bandwidth of 20 MHz. This is an example of wireless bandwidth. Looking at network performance bandwidth, one would identify it as the following: the network WAN connection only has a bandwidth of 1.5 megabytes (MB).

1.2 Wireless Signal Propagation When radio waves travel in the air, many things affect their quality, thus prohibiting them from actually transmitting their intended signals. Interference is one of the oldest and most difficult problems facing every type of wireless communication. This interference has caused such a design



4


Figure 1.1 Reflection.

challenge throughout history that many governments from around the world have had to step in to make certain frequencies restricted from use. Restricting this use prevents interference caused by other wireless devices and makes for cleaner airwaves. What happens to radio waves when interference affects their direction, influencing their signal clarity? Depending on what caused the interference, different common effects can occur. When the interference consists of certain objects, there are a number of well-documented, specifically proven results. When radio waves hit an object, they will bounce just like a child’s ball. They also have the ability to pass through some objects just as a ghost would. Being able to understand when each of these occurrences takes place is critical to understanding the operation of wireless.

1.2.1 Reflection Reflection takes place when an electromagnetic wave impacts a large, smooth surface and bounces off. This can happen with large surfaces such as the ground, walls, buildings, and flooring. After reflection takes place, radio waves often radiate in a different direction than originally intended. As one can see in Figure 1.1, the signal has a main pathway that intersects with the object. As it hits the object, it bounces off and heads in a different direction. This reflecting action lowers the signal




5

Figure 1.2 Refraction.

strength as it bounces off objects. Predominantly, the signal will pass through an object rather than bounce off of it. Reflection is one of the least obstructing interference types. This is because, for the most part, the signal remains whole; however, it moves in a different direction after it is reflected. Moreover, some of the other types of interference types will severely impact the signal’s quality.

1.2.2 Refraction When a signal reflects off an object and passes through it at the same time, one obtains what is called refraction (see Figure 1.2). RF is very stubborn; it goes places one does not want it to. Walls, buildings, or floors that should reflect the signal often do not; RF waves have a tendency to penetrate these objects instead. Once the signal has penetrated through these obstacles, it now has a degraded signal strength, which prevents it from reaching as far as it could have before the refraction. This is why reflection is not as bad an inherent interference as refraction. When a signal is reflected, most of the signal quality and strength is reflected with it. Refraction takes place when the signal has a portion of it penetrating and a portion of it reflecting. When this happens, the quality and strength are greatly deteriorated.

1.2.3 Diffraction Diffraction, which is similar to refraction, describes what a signal does when it encounters an object. In diffraction, after the signal makes it © 2006 by Taylor & Francis Group, LLC


6


Shadow Area

Figure 1.3 Diffraction.

around the object, we often get a shadow area. This is because the signal will bend around objects as best it can; but without being able to penetrate through the object, there is a dead spot created directly behind the object. Diffraction, unlike refraction, describes how the signal beams around objects instead of passing through them. People tend to get the two confused. In diffraction, shadow areas are created when an object will not allow refraction to occur. To picture this, see Figure 1.3, which shows the signal bending around the object; in doing so, it creates a shadow area directly behind the object. If refraction took place instead of diffraction, then the shadow area would not exist. This is because with refraction, the signal would bleed through the object and be present directly behind it. Some of the confusion around diffraction and refraction has to do with receiving a signal directly behind an object that the signal cannot penetrate. There are cases where this is true. It is possible for a signal to be unable to refract through an object but still be able to reflect enough times between different objects to make it around the main object.

1.2.4 Scattering Scattering (Figure 1.4) occurs when the RF signals encounter a rough surface or an area with tightly placed objects. The best way to understand scattering is to think of an automobile assembly line. In this scenario, one would see large amounts of robotic arms, raised metal-screened catwalks, pallets of metal doors, and many other objects. All these objects make the signal split into smaller signals, reducing the original signal’s strength. The main signal enters this area and reflects off the small metal objects and ping-pongs, thus creating more and more signals. Over time, this makes the main signal so scattered that its original strength diminishes. This is because when scattering takes place, the signal is equally divided among the many waves bouncing around the tightly packed area. On top of the signal strength reduction, this type of interference can cause © 2006 by Taylor & Francis Group, LLC



7

Figure 1.4 Scattering.

problems in receiving a signal. This is due to the fact that when multiple signals arrive at the receiver at the same time, it is difficult to correctly understand either of them.

1.2.5 Absorption Just by the name, one can probably figure this one out. When a signal hits certain objects — mostly water-based objects such as trees, cardboard, or paper objects — the RF signal actually is absorbed into the object. This one interference problem plagues point-to-point or point-to-multipoint bridge operations. Trees having a large amount of water in them tend to absorb large amounts of signals trying to pass through them. Evergreen trees are the worst because they store the most water inside them. When troubleshooting RF, beware of any large amounts of water-based products, objects, or stock. It often occurs that someone moves large amounts of palletized cardboard boxes and RF signals in that area diminish because of absorption.

1.3 Signal-to-Noise Ratio Within wireless networks, many types of interference exist. Some may be avoidable and other types are always present. The type of interference © 2006 by Taylor & Francis Group, LLC


8


that is always present stems from the movement of electrons and the basic radiation of energy. This means that no matter what one does, there will always be a slight amount of interference present in any airspace. This small level of interference makes up what is called the “noise floor.” To send a wireless signal, one must be able to transmit a signal above the noise floor. Once this is accomplished, one must overcome another interference type called “impulse noise.” Impulse noise consists of irregular spikes or pulses at high amplitude in short durations. This kind of interference can be caused by a number of things, ranging from solar flares and lighting to microwaves and walkie-talkies. The signal-to-noise ratio (SNR) helps wireless designers identify the quality of their transmissions. This is done by taking the signal power and dividing it by the noise power, producing the SNR value. This value is often measured in decibels (dB). The SNR value can help an RF designer understand how far the wireless area of coverage extends. In thinking about this, we are commonly under the mindset of increasing the power above the noise to fix our problems. Although this may be true, the FCC or, outside the United States, other government bodies regulate the amount of power a radio device can emit. However, this can impede one’s ability to easily get around interference issues by increasing power. The main goal of the government’s regulation is to prevent the basic radiation of energy from propagating out of proportion. If this was to happen, it would just increase the general noise floor for everyone, making it even more difficult to avoid interference. Looking at SNR values, one needs to understand a couple of facts about different values. First, an SNR value of 3 dB is equal to 2:1, which means that the noise level is about half that of the original signal. This number doubles for every 3-dB SNR value; this means if 3 dB is 2:1, then 6 dB is 4:1. Another fact is that for every increase of 3 dB, not only does one see the noise ratio change, but one also sees that the original power level has doubled. Using surveying tools, one may find oneself losing the connection around 5 to 9 dB. This is because one is getting very close to the 2:1 noise ratio explained previously. Most surveyors use a much higher value to take in account the different power types of wireless adapters and the movement of any interfering objects, such as stock on shelves. This value strongly depends on the environment and can fluctuate from 12 to 17 dB, giving an SNR value of 20:1 on the low end and 80:1 on the high end.

1.4 Modulation This section discusses some common modulation techniques so that one can get an understanding of how they work. Subsequently, this section



Wireless Network Overview 0

1

1

0

1

9

0

Figure 1.5 Basic modulation.

discusses some of the modulation techniques used by wireless networks. Before getting into the many types of modulation used on wir eless networks, one must understand what modulation is and how it works to increase bandwidth on a link. When discussing modulation, one must first focus on bits and baud and how they compare with one another. Bits, which are expressed as bit rates or typically related against time as bits per second (bps), are the measurement of data throughput. Baud is the rate of signal changes needed to send bits down a signal path. When one wants to take data and send it down a type of media such as a telephone line, it must be modulated into two different signals, which can be identified as a one (1) or zero (0). To do this, an oscillating wave is modulated by any number of techniques, such as amplitude, frequency, or phase, to create differences in the signal that can be received and returned to bits. Just like modems, wireless networks use modulation techniques to achieve communication and increase bandwidth. Looking at Figure 1.5 shows how an analog signal can be used to convey a one or zero, or vice versa. Exploring modulation gives a good idea about how wireless networks are able to jump in bandwidth just by changing their modulation technique. It will also help us understand how wireless networks actually send information. Using modulation techniques to increase bandwidth was also seen in the rapid increase of bandwidth on modems in the late 1980s. The modem designers found better ways to modulate the data and thus increase their throughput. Before starting the modulation, one needs to make sure there is an open communication channel. A carrier signal is what is used to ensure that the communication channel is open and modulation can take place. The awful sound a modem makes is its carrier signal connecting the transmitter and the receiver together before they start modulating data. © 2006 by Taylor & Francis Group, LLC


10

Wireless Security Handbook 0

1

1

0

0

1

Figure 1.6 Amplitude modulation.

1.4.1 Amplitude Modulation Amplitude modulation (Figure 1.6) is most often recognized in AM radio. This was one of the first and most basic approaches to modulation. It works by taking the signal and applying voltage to it to indicate the presence of data. When voltage is present on the line, it means a onebit notation or “on”; and when voltage is not on the line, it indicates a zero bit notation or “off.” Some coding mechanisms of amplitude modulation call out what is called a non-return to zero (NRZ); this means that if succeeding binary ones are present, the signal will continue to supply voltage for the given period of all the succeeding binary ones.

1.4.2 Frequency Modulation Frequency modulation (Figure 1.7), which most people use to listen to their favorite radio stations, is another modulation technique. Another name for frequency modulation is frequency shift keying (FSK); this comes from the old telegraph system wherein the operator would key in Morse code to relay a message. To understand how frequency modulation works, let us look at the old telegraph system. When an operator was waiting for a message to be sent, the key on the telegraph system was not pressed and no signal was going down the line. Once someone wanted a message to be sent, the operator would push Morse code onto the key and each time a signal would be sent down the line. This change in frequency was from no frequency to a frequency. Once the telegraph became automatic, a signal was always present; and once each key of the message was pushed, the signal changed to a higher frequency, giving us frequency modulation. © 2006 by Taylor & Francis Group, LLC


Wireless Network Overview 0

1

1

0

0

11

1

Figure 1.7 Frequency modulation. 180° 270°

0°

360°

90°

Figure 1.8 Phase modulation.

1.4.3 Phase Modulation Phase modulation is the one of the more common modulation techniques in use today. This is because it has the greatest ability to carry data when compared to the other modulation techniques we have looked at. Phase modulation has many different flavors itself. Some of these flavors incorporate the dual use of phase modulation and the previous techniques looked at in this chapter. A basic definition of phase modulation is the process of encoding information into a carrier wave by varying its phase in accordance with a type of input signal. Looking at Figure 1.8 provides a basic understanding of this. If one looks at a carrier wave, in this case a simple sine wave, one can see that its starting point corresponds to 0°. When the wave peaks, one has 90°; as it retunes to zero, one does not call it zero, but rather 180° because it returned from 90° and one can differentiate it from a wave just starting at 0°. In addition, one can also use the negative portion of the wave. As it reaches its negative peak, one has 270°; when it returns to zero, one has 360° instead of zero because it came from the negative peak. Now, to phase this sine wave, one needs to delay the wave’s cycle. In doing this, one can see that the wave should be at 180°, when in effect it is at 270°, making it 180° out of phase. © 2006 by Taylor & Francis Group, LLC


12


Now that we understand phase modulation, let us see how it is used to encode data. One of the simplest ways for phase modulation to encode data is called binary phase shift keying (BPSK) modulation. In this technique, one uses a simple 0° phase change that equals a binary 0 and 180° phase change that equals a binary one. When the signal is sent without any phase changes, it represents a binary zero; when there is a change, one will see a 180° change in phase, which repents a binary one. This can be increased using the other degree markers such as the 90° marker and 270° marker. When all four phase change degree markers are used, one has what is called quadrature phase shift keying (QPSK). One can also introduce a more angular phase change; however, the more closely the phase change gets to another, the more difficult it is to distinguish the size of the signal’s phase change. In direct relation to wireless networking, there are some modulation methods to look at. The first is included in the 802.11 standard and is called differential binary phase shift keying (DBPSK). This method is similar to the binary phase shift keying (BPSK) discussed above. It uses 180° of phase change to repent a binary one and 0° of binary change to repent a binary one. This means that if the data that must be sent is 0010, the wave’s signal will flow as follows. The first two zeros would be sent and no phase change would take place. Once the binary one was set to be transmitted, the phase would change to 180° out of phase. This would represent a binary one. After that, the signal would return to zero phase change, which indicates that binary zero was transmitted. The DBPSK produced the 1-MB data rate in wireless 802.11. As we will see in Chapter 6, the 802.11 standard was capable of producing a 2MB data rate. To achieve this, another modulation technique was used, called differential quadrature phase shift keying (DQPSK). This technique is used by a number of cellular technologies as well as the 802.11 standard. It is very much like the quadrature phase shift keying (QPSK) discussed previously. It works by having four points of reference for phase change. So, the 0, 90, 180, and 270 were used to allow encoding of more binary bits.

1.4.4 Complementary Code Keying (CCK) Once the 802.11b standard was released, another modulation method called complementary code keying (CCK) was included to reach higher data rates. This method uses QPSK in a similar fashion, although it employs coding techniques to increase the coding. It is performed by a complex mathematical symbol structure that repents encoded binary bits. These symbols can endure extreme interference levels and have very little chance of being mistaken for each other.




13

1.4.5 Quadrature Amplitude Modulation (QAM) When looking at modulation techniques, one sees the three discussed thus far in this section. Another method that has come out involves using two of these methods together. When one puts phase modulation and amplitude modulation together, one gets quadrature amplitude modulation (QAM). This is a technique in which both the phase and amplitude of a carrier wave are varied to allow for even more data bit encodings. In this, one not only has up, down, left, and right attributed to degrees and code bit, but also different levels of amplitude that allow for more bits to be encoded and sent. The 802.11a and 802.11g standards outlined in Chapter 6 used a technique called orthogonal frequency division multiplexing (OFDM). Inside the OFDM standard are four types of modulation techniques: (1) binary phase shift keying (BPSK), (2) quadrature phase shift keying (QPSK), (3) 16-quadrature amplitude modulation (16-QAM), and (4) 64quadrature amplitude modulation (64-QAM). Having discussed the first two modulation types, one can now look at the latter two: 16-QAM and 64-QAM. Instead of using the technique discussed above, the OFDM standard took on a different approach. They used the signal constellation and broke it into four parts. Imagine an X- and Y-axis crossing to obtain a cross; inside the cross there are four distinct sections, which are used by 16QAM to represent four subsections inside each of the original sections. This is illustrated in Figure 1.9 where one can see each of the four sections and the subsections. To change from 16-QAM to 64-QAM, one would use 6 encoded bits instead of 4 and 64 locations instead of 16. Digital television is one example of 64-QAM technology.

1.5 Wireless Groups When discussing wireless groups, two main categories come to mind: (1) the wireless governing bodies on a national and international level, and (2) the bodies that create interoperability standards to promote standardization of technologies. This section outlines both of these groups and looks at how they were created, why, and what benefits they provide. Looking closer at the first group of wireless bodies, one notices that these groups exist on a national and international level. This is because the threats of interference and the goal of creating worldwide wireless networks have always applied to each country in the world. Because of this, a global wireless body was created. Now, going back to each region or nation, one has small groups that detail the exact usage of the spectrum



14


0010

0110

1110

1110

3

2 0011

0111

1111

1011

1 −3

0001

−2

−1

0101

1

−1

2

3

1101

1001

1100

1000

−2

0000

0100

−3

Figure 1.9 16-QAM.

that is globally allocated by the international groups. Reading further into this section will provide an outline of how this works and how each of these bodies works with the other to prevent radio spectrum chaos. The wireless industry started out with vendors designing and creating their own wireless solutions. This made each network proprietary to that vendor; and if a vendor went out of business, so did any ability to get more of the needed network equipment. Wireless groups were created to make wireless technologies better able to interoperate between multiple vendors. The creation of wireless groups led to decreased time to market for new products, as well as more interoperability between vendors. These wireless groups, or standards bodies, create the main guidelines that wireless networks must follow. These groups have many internal problems that come about between what each manufacturer thinks is right; but overall, it is much better than it was before any wireless groups existed.

1.5.1 International Telecommunications Union (ITU) The International Telecommunications Union (ITU) was formed on May 17, 1865, in Paris, France. The reasoning behind this union was to streamline the process by which telegraphs were sent internationally.




15

Before this union, each county had expended time and resources fulfilling the requirements of each independent country. The complexity of dealing with each country and each of their requirements led to a meeting to address this issue. In this meeting, which lasted two and a half months, the ITU was created. This allowed each of the participating world governments to meet and create, agree, and modify different methods of communication. This union had 20 founding members when it was first created. On October 15, 1947, the ITU became a specialized agency under the United Nations (UN). During this time, the ITU created the International Frequency Registration Board (IFRB) to handle the task of managing the radio-frequency spectrum. This group was in charge of the Table of Frequency Allocations, which accounted for all frequency spectrum use throughout the world.

1.5.2 International Telecommunications Union Radio Sector (ITU-R) International Telecommunications Union Radiocommunication Sector (ITU-R) is a sub-group created by the International Telecommunications Union. The ITU-R is in charge of the technical characteristics and operational procedures of all wireless services. As part of its charter, the ITUR develops and maintains the Radio Regulations. This regulation serves as a binding international treaty that governs the use of the radio spectrum for all of its members worldwide. One of the key documents that the ITU-R is in charge of is the Radio Regulations. This document is a subsection of The International Frequency Registration Board’s Table of Frequency Allocations. The Radio Regulations were created in 1906 in Berlin, Germany, and address the frequencies ranging from 9 kHz to 400 GHz in the Table of Frequency Allocations. Today this document contains more than 1000 pages detailing how the spectrum can be used and shared around the globe. Making changes to this document is only allowed at a world radiocommunication conference such as the World Administrative Radio Council (WARC). During this event, members discuss, create, and ratify definitions for frequency allocation.

1.5.3 Federal Communications Commission (FCC) The Federal Communications Commission (FCC) is a United States Government agency established by the Communications Act of 1934. Its main goal is to regulate interstate and international communications. These communications include radio, television, wire, satellite, and cable. The © 2006 by Taylor & Francis Group, LLC


16


section of the FCC that deals with wireless technologies is the Wireless Telecommunications Bureau (WTB). Its service includes cellular telephone, paging, personal communications services, public safety, and other commercial and private radio services. The WTB is also the bidding authority for spectrum auctions. The main goals of the Federal Communications Commission’s Wireless Telecommunications Bureau are to: Foster competition among different services Promote universal service, public safety, and service to individuals with disabilities Maximize efficient use of spectrum Develop a framework for analyzing market conditions for wireless services Minimize regulation, where appropriate Facilitate innovative service and product offerings, particularly by small businesses and new entrants Serve WTB customers efficiently (including improving licensing, eliminating backlogs, disseminating information, and making staff accessible) Enhance consumer outreach and protection and improve the enforcement process

1.5.4 Conference of European Post and Telecommunications (CEPT) The European Conference of Postal and Telecommunications Administrations — CEPT — was established in 1959. At that time, only 19 countries were involved. As the CEPT gained momentum, it was able to expand into 26 countries within its first 10 years. After 29 years of operation, the CEPT organization decided to create the ETSI, the European Telecommunications Standards Institute. The ETSI was created to handle standardization and not regulatory issues such as spectrum allocation. To get a better understanding of the CEPT and ETSI, one can compare them to the United States and its creation of the Federal Communications Commission (FCC) agency and the Institute of Electrical and Electronics Engineers (IEEE). The FCC manages issues concerning spectrum allocation and power regulations inside each allocated spectrum. The IEEE builds standards for devices that can operate in one of the allocated spectrums. The members of the CEPT as of September 21, 2004, are listed as follows:




17

Albania, Andorra, Austria, Azerbaijan, Belarus, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Great Britain, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, Moldova, Monaco, The Netherlands, Norway, Poland, Portugal, Romania, Russian Federation, San Marino, Serbia and Montenegro, Slovakia, Slovenia, Spain, Sweden, Switzerland, the former Yugoslav Republic of Macedonia, Turkey, Ukraine, Vatican

1.5.5 Wi-Fi Alliance The Wi-Fi Alliance is a nonprofit international association formed in 1999. Its main goal is to certify the interoperability of wireless local area network (LAN) products based on the IEEE 802.11 specification. Wi-Fi stands for wireless fidelity. The Wi-Fi Alliance has certified more than 1000 products with its Wi-Fi® certification. This association came about due to the lack of well-defined technical areas in the 802.11 standard. As seen later in this book, most of the wireless standards lack certain details. For example, the 802.11 standard states that roaming will be supported, although it does not detail how a manufacturer should allow for roaming. This means that the Wi-Fi Alliance will only certify products to what is defined in the standard. The security mechanism called WEP only started as a 40-bit key in the original 802.11b standard. In the security section of this book, one sees that the key size of WEP was increased to 104 bits; this was done outside the IEEE standard. This means that for the Wi-Fi Alliance to certify a product, it only has to support a 40-bit key rather than the more often recommended 104-bit key. The Wi-Fi Alliance’s goal was to make sure that if a product is Wi-Fi certified that it would interoperate with other Wi-Fi-certified products. The original name of the Wi-Fi Alliance was the Wireless Ethernet Compatibility Alliance (WECA).

1.5.6 IEEE The IEEE, which is an acronym for Institute of Electrical and Electronics Engineers, is the group that created all the 802 standards. This also includes the wireless standards in the 802.11 space. The IEEE has been around since 1884 although it was not always called the IEEE. In 1963, the AIEE (American Institute of Electrical Engineers) and the IRE (Institute of Radio Engineers) merged. This came about from the existence of two separate standards bodies that were made up of many of the same people. Instead



18


of them arranging two different meetings with each other for very similar objectives, they decided to merge the two organizations. Many brilliant minds, including Thomas Edison, were part of the AIEE, which is now known as the IEEE. The IEEE is a governing body that created the 802 standards for network communications. The requirements needed to create an IEEE standard have a well-defined process that has seven layers. These layers allow the standard to move from thought to a written, defined, and approved IEEE standard. The seven-step process is outlined as follows: 1. 2. 3. 4. 5. 6. 7.

Call for interest Study group Task group Working group ballot Sponsor ballot Standards board approval Publication

The process starts out with a call for interest in which the IEEE kicks off a meeting about a peculiar standard. In our case, this would most likely be a new wireless standard. The IEEE has a large scope, well over the small wireless subsection that relates to our example. Once the call for interest has been performed, a meeting will take place. In this meeting, attendees discuss the need for this type of standard and whether or not it is even needed. Depending on how they react to this initial meeting, the IEEE can continue with this standard or can stop it here. If they decide to continue, the next step is to develop a study group of participants to look into this further. This group would work together to discuss and decide if they are willing to commit to the next phase, in which a standard will be drafted. Once the study group moves to the task group, they are going to start writing the standard draft. Once the draft is finished, it will need to go to a working group ballot. At this point, the standard must receive a 75 percent approval rate until it can move to the next step. Most of the time, many drafts are created in this phase until one finally receives the required votes. This is when the most disagreement and time-consuming discussion takes place. Frequently, this is because each vendor involved with a particular standard has already invested R&D dollars into something that another group member may want to change. This political battle takes place until a vote reaches the 75 percent mark needed to move to the next step. After the vote meets the 75 percent mark, it goes through another ballot in which executive members of the IEEE vote on it. After this phase, it goes to the IEEE to approve and publish.




19

1.6 Chapter 1 Review Questions 1. What happens to an 802.11b wireless signal when an evergreen tree is located between the transmitter and receiver? a. Nothing. b. It is refracted. c. It is diffracted. d. It is reflected. e. It is absorbed. 2. What are the two correct terms used to measure antenna gain? a. dBi and DBd b. Watts and milliwatt c. dBd and dBi d. EIRP 3. The designator dBi is a decibel compared to what? a. Milliwatt b. Decibel c. Isotropic radiator d. RADIUS 4. What does RF stand for? 5. Which type of modulation does 802.11b use? a. QAM b. FM c. AM d. CCK 6. How can one send more data across the air? a. Increase the transmit power. b. Use a more complex modulation. c. Use a bigger antenna. d. Use a wider frequency band. 7. What was the Wi-Fi Alliance formerly known as? a. FCC b. WECA c. IEEE d. WIFI



20


8. What seal certifies interoperability in a manufacturer’s wireless device? a. WHY b. Hi-Fi c. WECA d. Wi-Fi Certified e. Wi-Ki 9. Which wave will travel the greatest distance? a. FM radio b. X-ray c. 802.11a d. Microwave 10. What two items should be maintained near the edges of a wireless cell when performing a site survey? a. High signal-to-noise ratio b. Low signal-to-noise ratio c. High noise level d. High signal strength 11. What would the FCC and ETSI regulate on a wireless network? (Select more than one) a. Power outputs b. Total client number c. Channel number and frequency d. Who can use the system 12. What bandwidth term is this phrase stating? On any given day, my wireless network has a low bandwidth of ______. a. 11 Mbps b. 2.4 GHz c. 11 MHz d. 5.4 GHz 13. Which of the following show the correct use of a wireless network? a. Using wireless to connect two buildings point-to-point b. Mobile access from laptop or PDA c. As a way to connect a server d. To increase bandwidth on a 10/100 wired network




21

14. When a wireless signal changes or bends around an object, sometimes creating a shadow area, it is known as __________. a. Refraction b. Reflection c. Diffraction d. Scattering 15. When performing wireless power calculations, what two terms are often converted into each other? a. dBi to dBi b. dB to Watts c. dBm to DBi d. dBd to dBi 16. What standards body creates wireless standards? a. Wi-Fi Alliance b. IEEE c. FCC d. WECA 17. What key size is required in WEP for the Wi-Fi Alliance to certify a product? a. 40 bits b. 120 bits c. 128 bits d. 56 bits 18. What does Wi-Fi stand for? a. Wireless infrastructure fidelity industry b. Wireless Interoperability Forum Institute c. Wireless fidelity d. Wireless networking



Chapter 2

Risks and Threats of Wireless This chapter discusses the general goals for information security and how they are used to measure risk and understand threats. This information will help in the next sections of this chapter where each of the threats relating to the many types of wireless communications is explored. After looking at each of the threats, this chapter focuses attention on wireless hackers. In this chapter, we see how hackers locate the existence of wireless networks as well as how law enforcement tracks down these hackers.

2.1 Goals of Information Security When looking at information security, one must address the three tenets of information security: (1) confidentiality, (2) availability, and (3) integrity. These long-standing goals will help us understand what we are trying to protect and why. This information will help when one starts looking at all the risks and threats that face wireless communications. Before one can properly evaluate risk, one needs to set a baseline to understand the definition of each goal one is trying to uphold.



24


2.1.1 Confidentiality Attacks on the confidentiality of information relate to the theft or unauthorized viewing of data. This can happen in many ways, such as the interception of data while in transit or simply the theft of equipment on which the data might reside. The goal of compromising confidentiality is to obtain proprietary information, user credentials, trade secrets, financial or healthcare records, or any other type of sensitive information. Attacks on the confidentiality of wireless transmissions are created by the simple act of analyzing a signal traveling through the air. All wireless signals traveling through the air are susceptible to analysis. This means there is no way to have total confidentially because one can still see a signal and subsequently record it. The use of encryption can help reduce this risk to an acceptable level. The use of encryption has its own flaws, as seen later in this book. For the most part, the encryption is secure itself, although how it is implemented and how key management is handled may produce flaws that are easily exploited.

2.1.2 Availability Availability is allowing legitimate users access to confidential information after they have been properly authenticated. When availability is compromised, the access is denied for legitimate users because of malicious activity such as the denial-of-service (DoS) attack. Receiving RF signals is not always possible, especially if someone does not want you to. Using a signal jammer to jam an RF signal is a huge problem that has been facing national governments for years. Looking for the availability of RF local area networks (LANs), one notices that performing a DoS attack is easy to accomplish. This is due to the low transmit power allocated by the U.S. Government and poor frame management techniques included in most of the current day wireless standards.

2.1.3 Integrity Integrity involves the unauthorized modification of information. This could mean modifying information while in transit or while being stored electronically or via some type of media. To protect the integrity of information, one must employ a validation technique. This technique can be in the form of a checksum, an integrity check, or a digital signature. Wireless networks are intended to function in an unimpaired manner, free from deliberate or inadvertent manipulation of the system. If integrity is not upheld, it would be possible for an attacker to substitute fake data.



Risks and Threats of Wireless

25

This could trick the receiving party into thinking that a confidential exchange of data is taking place when in fact it is the exact opposite. Wireless networks have adapted to this type of threat over time. One can see this advancement as new security standards emerge, creating increasingly complex integrity checks.

2.2 Analysis Analysis is the viewing, recording, or eavesdropping of a signal that is not intended for the party who is performing the analysis. All RF signals are prone to eavesdropping; this is because the signal travels across the air. This means anyone within the signal’s path can hear the signal. One of the only protections available to prevent the loss of confidentiality is encryption. If a signal is using encryption, then its confidentiality can be upheld until that form of encryption is defeated. The risk of analysis on an RF signal is an inherent risk that cannot be avoided. The only option is to mitigate the risk with some type of confidentiality control.

2.3 Spoofing Spoofing is the act of impersonating an authorized client, device, or user to gain access to a resource that is protected by some form of authentication or authorization. When spoofing occurs in wireless networks, it primarily involves an attacker setting up a fake access point to get a valid client to pass authentication information to that attacker. Another way attackers spoof is by performing a man-in-the-middle attack. In this scenario, an attacker would position himself between a client and the network. This could be accomplished by spoofing a valid access point or by hijacking a session. Once this part is complete, the attacker would then use the authentication information provided by the client and forward it to the network as if it originally came from the attacker.

2.4 Denial-of-Service Denial-of-service (DoS) is the effect of an attack that renders a network device or entire network unable to communicate. Hackers have found that certain crafted packets will make a network device unresponsive, reboot, or lock up. They have used this technique to shut down hightraffic networks and Web sites. They have also used this attack to reboot network equipment in an attempt to pass traffic through the device as it



26


is booting up. This is done to try to circumvent any policies set up on the device to protect it or devices behind it. The DoS threat can also adversely affect the availability of a network or network device. Wireless DoS attacks can be achieved with small signal jammers. Finding signal jammers is not as difficult as one might think. Some modernday wireless test equipment can perform jamming. This is not the tool’s intended purpose, although it is commonly used for this. Jamming is possible because the government regulates the amount of power allowed on a wireless network. In relation to wireless LANs, the amount of power used is a very small amount. This means that it is not difficult to overpower an existing device with a home-made one. Another DoS threat relating to LANs in particular is the poor structure of management frames. These frames allow for anyone who can analyze the wireless signals to perform a DoS attack by replaying certain management frames. Mostly, theses attacks are layer two frame attacks. These attacks try to spoof management traffic, informing the client that he is no longer allowed to stay connected to the network. Chapter 13 discusses these attacks in more detail.

2.5 Malicious Code Malicious code can infect and corrupt network devices. Malicious code comes in many forms: viruses, worms, and Trojan horses. People often confuse the three main forms of malicious code. Because of this, they use these terms interchangeably. This section looks at each of these and identifies what classifies them into each of the three groups. Viruses infect devices and do not have the ability to replicate or spread outside the infected system on their own. Once a virus infects a machine, it can only replicate inside the infected machine. This means that all threats from viruses stem from receiving infection. The threat of worms is much higher because they can spread across the enterprise and out to the Internet, infecting multiple devices. In the past few years, humans have started to see global worms that propagate across the entire world. The final malicious code threat discussed here is the Trojan horse. This threat comes from installing or running programs that can have or within their use execute code that might contain malicious content. Malicious code relating to wireless has to do with new viruses that can affect the many new types of wireless end devices such as PDA units, smart phones, PDA phones, laptops, etc. Wireless viruses have just started to appear in the wild. Even with this threat just starting to develop, many forms of wireless malicious code have already appeared. Some of this code has enough intelligence to find and utilize a variety of available wireless technologies on a device to spread even further. © 2006 by Taylor & Francis Group, LLC



27

Another form of malicious content relating to wireless is spam. Although spam is not destructive in nature, the time and money it costs an organization often makes it seen as malicious. Spam is not just related to wireless. Long before wireless spam there was e-mail spam. Today’s wireless devices are capable of receiving messages in many formats: email, text messaging, instant messaging, and voice calls. All of these are starting to see spam pop up on them. Dealing with spam has created a security market of its own with products, solutions, and services created to combat this threat.

2.6 Social Engineering Social engineering is the often called low-tech hacking. It involves someone using the weakness of humans and corporate policies to obtain access to resources. Social engineering is best defined as tricking or manipulating a person into thinking the party on the phone is allowed access to information, which they are not. The threat of social engineering has been around for quite some time. Some of the most well-known computer hackers used this type of attack to get information. The real threat to this is the skill level involved. No one needs to be computer savvy or a technical genius to perform this type of attack. There are a number of things to do to prevent this type of attack. First, make sure that a policy is in place regarding sensitive information and phone usage. Make sure that not anyone can call and reset someone’s password. Create a helpdesk identification process to authenticate callers to the help-desk operators.

2.7 Rogue Access Points Rogue access points pose a major threat to any organization. This is because of the high availability and the limited security features of offthe-shelf access points. If a company does not approach the WLAN (wireless local area network) concept fast enough, frustrated employees will take it upon themselves to start the process. When this happens, employees often put in wireless systems of their own. Even with most current-day access points supporting advanced security standards, the default configuration of an out-of-the-box access point is set to the least secure method. This has created a real threat because now a user can easily bring in a rogue access point, plug it in, and put the entire network at risk. The knowledge level required to install an off-the-shelf access point has almost become plug-and-play today. This means that more and more people have the ability to place rogue access points. These same



28


people lack the ability to secure these devices or even understand the risk they are posing for the company. Most access points come from employees, although as we will learn later there are cases where an attacker would try to set one up for easy return access. This was not a big issue until recently when the price of 802.11b access points fell well below $100. To do this, an attacker would need physical access and a network port. If a hacker wanted access bad enough, spending $100 for it would be a conceivable expense. With companies investing in stronger security mechanisms, it would be a shame to have an incident in which an attacker gains access through a non-secure rogue access point. Because of the threats associated with rogue access points many companies have started to put controls in place to increase awareness and prevent the deployment of rogue access points. Many companies that jumped into the newly formed wireless security market have adapted and created tools to detect rogue access points. Some companies have handled rogue access points by creating policies about wireless usage and strict penalties for rogue access placement. Others have taken a second route and invested in wireless intrusion detection systems (WIDS) software.

2.8 Cell Phone Security Now we will have a discussion of general cell phone identification and security. Cell phones have had a slight advantage over other types of wireless communications in the security realm. This is due to their overwhelming numbers. Most people today have a cell phone; and with so many people using cell phones, many security risks and subsequent controls have been developed to counter each other. Understanding this information will show how cellular phone providers have mitigated similar risks that face wireless local area networks. Cell phones send radio frequency (RF) transmissions on two distinct channels: (1) one for actual voice communication and (2) the other for control signals. This control signal identifies itself to a cell site by broadcasting its mobile identification number (MIN) and electronic serial number (ESN). When the cell tower receives the MIN and ESN, it determines if the requester is a legitimate user by comparing the two numbers to a cellular provider’s subscription database. Once the cellular provider has acknowledged that the MIN and ESN belong to one of its customers, it sends a control signal to permit the subscriber to place calls. Like all RF devices, cell phones are vulnerable to eavesdropping and spoofing. In the cellular phone industry, these are called “call monitoring” and “cell phone cloning.” Another risk associated with cell phones is the




29

ability to reprogram phones, transforming them into advanced microphones capable of recording and transmitting sound from their location to anywhere in the world. Monitoring calls is an easy task, especially for phones that use analog technology. This is because most analog cell phone technologies were transmitted in the same band as FM radio. A commonly available radio frequency scanner could get one up and listening to calls in minutes. With the proliferation of digital cellular networks, more and more security was erected. This was great because inside a service provider’s network, your calls were, for the most part, safe. There were easier analog targets for criminals to exploit. One’s digital phone was not so safe if one roamed or went outside of a provider’s area of coverage. When two cellular providers wanted to hand off calls to each other for billing purposes, they converted them to analog so they had a common protocol for interoperability. This also meant that security was no longer present. So, even with a digital phone, once the MIN and ESN are removed or identified from the phone call, it could still be tracked, cloned, or monitored inside the digital network. Another trick involves turning a cellular telephone into a microphone and transmitter. This can be used to record a conversation or bug a room. This can be done without your knowledge by police, governments, and even some highly educated people. How does it work? It is easy to do, just send a maintenance command on the control channel to the phone. This command places the cellular telephone in a diagnostic mode. When this is done, conversations in the immediate area of the telephone can be monitored over the voice channel. The signal engages the phone to perform this monitoring action without any indication of it taking place. The user does not know the telephone is in the diagnostic mode and transmitting all nearby sounds until he or she tries to place a call. The calling feature does not work and the phone is useless until the power is cycled. After that, the phone returns to a normal state as if nothing ever happened. This is very scary because the user has no idea he is bugged by his own phone through the airwaves. This threat is the reason why cellular telephones are often prohibited where classified or sensitive discussions are taking place. Someone could be bugging your phone as you read. Do not worry; as long as one can place a call without cycling the power, you’re ok. One publicized case of cell phone monitoring involved former Speaker of the House of Representatives, Newt Gingrich. A call between Gingrich and other Republican leaders was monitored and taped. The conversation concerned Republican strategies for responding to an ethics violation for which Gingrich was being investigated. This call was given, or most likely sold, to the New York Times and made public.



30


Another publicized case of cell phone monitoring involved a pager system instead of a cellular phone system. In 1997, the Breaking News Network monitored the pager messages of a large number of New York City leaders, including police, fire, and court officials. The messages recorded were considered too sensitive to send over the government’s protected police radio. This confidential information was captured and then sold to other news agencies in order to get the scoop on a story. This ended up happening sometimes before the police dispatch even had the information. Later in the year, police arrested the officers of this New Jersey news company for illegally monitoring their pager systems. Next we look at cellular phone cloning. What is cell phone cloning? It is the copying of the unique identification information programmed into your cell phone by a cellular provider. The cellular provider programs the phone with an electronic serial number (ESN) and mobile identification number (MIN). A cloner will steal this information, copy it to a different phone, and place calls on your bill. There are many ways for cloners to obtain these numbers. One is when someone fixes your phone or even when you buy a new phone at the store, someone could copy this information during the activation process. The MIN and ESN can also be obtained by an ESN reader (see Figure 2.1), which is similar to a cellular telephone receiver designed to monitor control channels. The ESN reader captures the MIN and ESN as they are being broadcast from a cellular telephone to a cell tower. This happens when your phone is turned on or when you move from one cell phone tower to another. A major controversy grew around cell phone cloning. At first, the phone companies refused to admit that their security was compromised, thus making the victim pay for all the calls placed by the cloner. This proved to be a big problem for cell phone companies and their customers. Another threat related to cellular phones deals with the short messaging service (SMS). This is a method of sending short messages similar to email. One of the threats related to this has to do with mass SMS messages that create a denial-of-service attack. This sort of attack has not been widely seen yet, although many industry leaders have openly spoken about the risk and impact if it were to happen.

2.9 Wireless Hacking and Hackers Inherently, RF has many threats, including interception, signal jamming, and signal spoofing. Because RF travels through the air, picking up the signal is as easy as being within the radio waves’ vicinity with the right hardware. Spectrum analyzers can detect radio transmissions showing the




31

Figure 2.1 ESN Reader.

user the signal frequency. Depending on that frequency, an attacker might be able to identify the transmission right away. Most RF frequencies in the spectrum are reserved for specific uses. Once one is able to find a signal and map it to a reserved spectrum, one knows who is transmitting it and, in some cases, why. Getting more in tune with the majority of RF threats, one can look at today’s RF local area networks (LANs). These, of course, have the same threats as all RF signals, although they do add a new dimension stemming from mass use and scrutiny. Just like cell phones, the more people who use them, the more time people spend looking at how they work and what they can do to defeat any security that exists. This has been seen over the past few years as a large number of users started deploying wireless networks and security flaws began to pop up. Most wireless network setups are capable of working right out of the box. This has led more and more nontechnical people to deploy them. When setting up a wireless LAN right out of the box, the default configurations are usually the most insecure ones.

2.9.1 Motives of Wireless Hackers America’s laws and law enforcement agencies have taught us that rarely is a crime committed without a motive. With this said, if someone was © 2006 by Taylor & Francis Group, LLC


32


to spend the time to compromise an RF signal, there always is a motive. Some of these motives can be as harmless as wanting an Internet connection to send a loved one an e-mail, or as terrible as committing an act of terrorism against a nation or government. To understand why someone would try to compromise an RF signal, take a look at some of the more well-known motives, such as to get a free Internet connection, commit fraud, steal sensitive information, perform industrial or foreign espionage, and — the worst of all — terrorism. After understanding what motive or motives an attacker might have, one can better understand how much security one should apply to the RF signal. If a company deals with financial information, it probably is more at risk from an attacker than a small doll shop. Knowing who might attack and why can help ensure that the correct risk-reducing actions are taken.

2.9.2 War Drivers After more and more people realized that out-of-the-box wireless LANs were generally set up, by default, in the most insecure mode, people started to exploit them. This new fad of identifying and categorizing wireless networks based on their security level has been coined “war driving” (see Figure 2.2). War drivers use equipment and software to identify wireless networks. War driving allows attackers to understand the security associated with any particular wireless network they happen to pass. This equipment not only allows war drivers to identify, locate, and categorize wireless networks, but also allows them to upload their results

Figure 2.2 War drivers.



Risks and Threats of Wireless 33

to a Web site where their friends and everyone else who has access will be able to see exactly where these unsecured wireless networks are located. This has even become so advanced that the use of GPS has been incorporated to give other people exact locations to these insecure networks. Anyone can simply go online and get a map of the exact location of an insecure network identified by a war driver. This culture has taken on its own members who are not stereotypical classic teenage computer crackers or hackers. Some older, wealthier people have begun to war drive. One individual turned his Acura in-dash GPS into a war-driving display. Others have mounted fixed wireless antennas onto their vehicles. Figure 2.2 shows a war driver using a Cadillac. Today, there are major events as well as Web sites dedicated to indexing and storing wireless network location information. They use information provided by war drivers, and some have gone as far as or ganizing worldwide war drives. The Web site www.worldwidewardrive.org has done just this for four years running. Their first worldwide war drive started on August 31 and ended on September 7, 2002. In this short period of time, 9374 access points were found; and of those, only 30 percent had any encryption technology. As time passed, the worldwide war drive went into round two, lasting from October 26 to November 2, 2002. In this round, they found 24,958 access points, with the number of unsecured access points rising 2.2 percent. With rounds three and four, the number of unsecured access points fell, although they only fell a small percentage. With the fourth worldwide war drive now complete, the total number of wireless networks running without any type of encryption is 61.6 percent. Figure 2.3 has a chart that shows the data collected from all four worldwide war drives. This information was obtained from the worldwide war drive Web site. The latest information available about this year’s or upcoming years’ worldwide war drives is also available on that site.

2.9.3 War Walkers A more athletic approach involves war walking instead of war driving. In this concept, a war walker would stroll down the street with a laptop either in a bag or out in the open. They would use the same tools and equipment employed by the war driver to identify insecure networks. This has gained a lot of steam, given that the number-one profile for a computer hacker is a teenager who most likely does not have a car or is not old enough to drive. Consumer industries have even gotten on this bandwagon by producing tools to find wireless networks. There are even devices that connect to your keychain that will beep or light up when wireless networks are detected. This has turned the act of war walking into an event that can be performed without any effort during any activity that requires © 2006 by Taylor & Francis Group, LLC


34


WWWD4 WWWD3 WWWD2 WWWD1 0.00%

20.00%

40.00%

60.00%

80.00%

WWWD1 WWWD2 WWWD3 WWWD4 Using Default SSID Without Encryption With Encryption

29.53% 69.86% 30.13%

35.27% 72.07% 27.92%

27.83% 67.74% 32.26%

31.40% 61.60% 38.30%

Figure 2.3 Worldwide war drive stats.

Source: www.worldwidewardrive.org. movement. This means that a war walker can perform this malicious activity while he or she gets milk for mom.

2.9.4 War Chalking Next is the concept of war chalking (Figure 2.4), in which a war driver or walker does not have the time to go to a Web site and locate an insecure network or does not have any available internet connection to do so. To help a fellow war driver or walker, the community has developed a number of symbols representing wireless networks and their associated security levels. This helps them to find the quickest, most insecure networks so that they can connect to the Internet and anonymously surf the Web or participate in any other activity on the Internet in an anonymous manner.

2.9.5 War Flying Interestingly enough, lately a new breed of identifying wireless networks has emerged, called war flying. As the name infers, war flying is the act of scanning wireless networks from inside the cockpit of a private plane. When a person is war flying, he can cover a large distance quickly. This has proven the quickest way to collect data about insecure wireless networks thus far. This is because of the use of private planes that can fly at a low altitude and cover a large distance rather quickly. The recommended altitude is less than 2500 feet for optimal wireless scanning. © 2006 by Taylor & Francis Group, LLC



35

SSID Open

Bandwith

Example

SSID

Linksys Closed

5.5 Meg

SSID W

WEP Protected

Bandwith

Figure 2.4 War chalking.

With war flying, we can see that this new trend of wireless scanning also appeals to people wealthy enough to purchase or rent a private plane.

2.9.6 Bluejacking Bluejacking is a relatively new term that focuses on Bluetooth-enabled devices. Unlike the name, bluejacking is not stealing or hijacking Bluetooth devices: rather, it is a way to send anonymous messages to Bluetoothenabled devices. A bluejacker will go to a place where there are a large number of people, such as on a subway, and scan the airwaves for Bluetooth devices. By default, many Bluetooth devices allow for pairing. When a bluejacker finds one of these devices, he sends messages to them and tries to identify them in the crowd. Once identified, messages that are more personal can be sent. This will generally make some nontechnical people a little scared. Could you imagine getting pop-up messages on your cell phone describing where you are and what you are wearing?

2.9.7 X10 Driving Lately, many X10 cameras are popping up in more than just your browser. With an inexpensive wireless camera out on the market, many people © 2006 by Taylor & Francis Group, LLC


36


are using them for all kinds of reasons. This has created what is now known as camera driving. Just as a Peeping Tom would look into your windows, these people buy X10 camera receivers and watch what is broadcasting on X10 cameras in any given area. Because these cameras have no security, any receiver can see what any camera is transmitting. There are other types of X10 devices available to automate a home. These devices can open the garage door, turn on and off the lights, and control some appliances. Similar to X10 cameras, these devices can be controlled by any receiver. This means not only can someone see what the X10 cameras are broadcasting, but they may also be able to get into one’s home or create problems with one’s lighting or appliances. Recently, X10 adopted a weak form of security called home and device codes. This security is applied by two 16-digit wheels that can be set to a house code and a device code. This means that only 256 possible codes exist. Although this might hinder the casual eavesdropper from easily picking up the devices, anyone who wants to take the half an hour to switch the code wheel between all 256 possible codes will be able to defeat the security and control any X10 device in that home.

2.9.8 Cordless Phone Driving When it comes to phone driving, most of this type of eavesdropping has been reduced by U.S. Government laws and enhanced security. Older 900-MHz phones were primarily analog and could easily be picked up by most scanners. With the advent of 2.4-GHz and 5.8-GHz cordless phones, many higher frequency scanners were more difficult to procure. In an attempt to preserve privacy, the U.S. Government made a large number of these scanners illegal to own or operate. This did not stop resourceful people from tapping into phone conversations, so phone manufacturers employed a security mechanism called Digital Spread Spectrum (DSS), which sends the call information in a digital format across multiple channels.

2.9.9 War Dialing We have discussed terms such as “war driving,” “war chalking,” and “war walking.” These new hacking terms originated from an old term called “war dialing.” These war dialers would dial multiple numbers looking for a modem connected to a computer system. This was because of the large number of insecure modems connected to many computers and computer networks. Applying this to wireless, there are now war drivers who drive around looking for wireless access points connected to networks. War




37

dialing and war driving are similar in the sense that the attacker is trying to connect to a predominantly insecure medium in hopes that the sheer number of deployments will result in poor security.

2.9.10 Tracking War Drivers After confronting a war driver once, I asked him, “Why do you do it?” His statement to me was rather interesting in the fact that it was a contradiction of terms, yet so many of his peers also had the same answer. One response was “to educate the public to the existence of these insecure means of accessing networks and the Internet.” The second response was “amenity or the ability to go online without any record of it being traceable back to them.” Now, thinking about this, if the goal was to have an untraceable Internet connection, then why expose the networks to the public eye? Well, at least the question was answered like a true politician. Credit must be given to a teenage hacker who has professional speaking skills like that. How would someone track down a war driver? The FBI had several public cases of arresting criminals using wireless networks to compromise retail store networks. They were tracked down so we know it is possible; so let us learn how. Once the investigation starts, a forensic team arrives on site and dumps the configuration and stored memory of all network devices and servers that were affected. Once this data has been properly removed, in accordance with the chain of evidence, it is properly examined at a lab. This examination process is a timely one, so much so that it can make many incidents considered not financially worth the effort. Many cases are too small to warrant the massive effort needed to investigate. After the lab results are examined, one can see where the perpetrator first entered the network. Because this was on a switch connected to an access point, one can determine that they came in over the airwaves. Once this information is identified, one can determine the wireless network interface card’s MAC address. This address is hard-coded onto the card by the vendor and is regulated in a sense, which makes it globally unique. Some clever hackers have the ability to change the card’s MAC address, but as time has shown, many do not take the time to do this. Chapter 13 (“Breaking Wireless Security”) discusses how this is done and what tools are out there to perform this type of attack. After the MAC address has been determined, one of two things can happen. First, the police can get a warrant to search any suspect’s home for the network card in hopes of finding it and its matching MAC address. In a highly important case, such as one that involves terrorism, the FBI



38


might go back to the card maker and track that card’s movement from creation at the manufacturer’s factory, to the distributor, then to the retail store and finally to the purchaser. This is an easy task to accomplish, although it is very time consuming. It works by correlating many different data sources to limit the number of people to question. Looking at the tracking of the card itself, a vendor can show proof of its arrival at a warehouse or retail store. Once it is proven that it has arrived at a retail store, one only needs to find out who bought it. The first round would be to look up all transactions on the point of sales machines for anyone who purchased any of the vendor’s network cards. Looking at this gives credit card information for anyone who used that method of payment. Most likely, if someone were going to do something illegal, he or she would have paid in cash. Well, it is also easy just to look on the store’s video camera correlating all the times that any of the vendor’s cards were purchased minus any purchases made with a credit card. After looking at how wireless war drivers can be tracked, one gets to a more important point about wireless devices. All bi-directional communicating wireless devices emit radio waves; so in a sense, all wireless devices can be tracked in one form or another. As one reads through this book, one will see that most modern-day wireless devices have some type of tracking method associated with them. Next time you see some amazing new RF technology, remember the statement above. No matter what manufacturers say about their technologies, any bi-directional communicating wireless device can be tracked.

2.10 RFID The Radio Frequency Identification (RFID) concept has created some major privacy issues. With RFID, companies can save time and money by being able to track products from their creation, to their purchase at a retail store, and beyond. It is the “beyond” part that has so many people upset about the inherent piracy issues of RFID. Before delving into those concerns, one needs an understanding of the technology. RFID systems have been used for quite some time. Only recently has their true potential been realized. An RFID system is a small tag that is affixed to an object to allow that object to be tracked. Once this tag has been turned on or energized, it will send information about itself when a reader queries it. This tracking can take place wherever there is a reader ready to query the tag. This means other companies can read RFID tags from their suppliers. There also is the ability to add to these tags; if one company buys a product from another and wants to insert some of its own data into the tag, they can. Another RFID innovation that has been discussed




39

prevents something almost all of us likely have done. How many people have ended up with a pink load of white laundry? Some washing machine manufacturers have talked about using RFID tags embedded inside clothing to prevent the red sock from getting into the load of white laundry. How about never reading tags for washing items again? The same manufacturers have talked about having washing machines set themselves based on the clothing item’s RFID tag. Most RFID systems today have a write-once tag, which means that erasing or modifying RFID information is unlikely. When the price of modifiable RFID tags come down in price, a new integrity threat will emerge, called RFID modification. Today, with RFID, companies can track their products and amass amazing amounts of data. There are so many ways to use RFID. Retail companies use RFID to per form automated inventory. Car manufacturers use them to tag special-order vehicles. Logistics companies use them to track package movement. There are even more usages that are created everyday; for an example of just how massive the push for RFID is, think about this: some companies have started to look at taking data from the RFID to feed financial reports, so investors know at any given time how many units were sold or shipped per quarter. This information could have a direct result on the price of stock. This could then, in turn, affect the way stock trading is performed in the future. Just imagine if a stockbroker could see in real-time the number of units a company is selling. Now that we have a good idea of how RFID works, let us look at the inherent risks and threats involved with its use. When discussing RFID, the first thing that comes to mind is the concern over privacy. In a world where the products one consumes transfer information to anyone willing to listen, the opportunity to market, trend, and collect data about us becomes a real concern. Some people have talked about many things relating to RFID, from the wild conspiracy theories to real issues that affect everyone on an every-day basis. To understand these concerns, one can look at a couple of examples that range from wild conspiracy theories to those that affect almost everyone every day. Using RFID, people with access to the right information assets can track individuals based on what they have purchased. If someone buys a can of soda from one store and then walks into another store, the second store’s readers might pick up that can. If someone wanted to find a person and had the resources, he or she could find the RFID tag ID number and cross-reference that with a credit card system or company database. This would allow a simple object to become a tracking device. This is highly inconceivable today; although inconceivable or not today, it is technically possible.



40


Another privacy concern that affects almost everyone is the ability to read and use information from product tags not belonging to the reader’s organization. To put this into context, imagine walking into a store with a bag or purse. When you walk in, card readers at the door energize all the items inside your bag. Then these items send all their information to the reader. The store now has a record of your purchases, not from that store but just in general. Everything inside your bag that has a UPC would have an RFID. These records could include over-the-counter medications, feminine products, and a number of other things that many people consider private. To make things worse, the salesperson might be given this information to get an idea about your purchasing habits.

2.11 Chapter 2 Review Questions 1. Which of the following processes will not lower the risk of social engineering at a help desk? a. Positively identifying the caller b. Using a callback method c. Shredding documents d. Having the caller verify the identity of the help desk operator 2. What would a hacker whose motive was money most prefer to attack? a. School b. Bank c. Doll shop d. Pizza shop 3. Which of the following terms best describes a Wi-Fi hacker? a. War dialer b. Hacker c. War hacker d. War driver 4. What type of malicious code infects devices and does not have the ability to replicate or spread outside the infected system on its own? a. Worm b. Virus c. Trojan horse d. Spam




41

5. List the three main goals of information security. a. Integrity b. Encryption c. Availability d. Confidentiality e. Scalability f. Protecting 6. What two pieces of information are required to hack a cell phone? a. MNN and ESS b. MIN and ESN c. ENS and MSN d. Phone ID and vendor ID 7. What piece of information is unique on every wireless card in the world? a. IP address b. Serial number c. MAC address d. SSID 8. Which of the following terms are used to describe the process of discovering wireless networks? a. War flying b. War walking c. War driving d. All of the above 9. A self-replicating and often self-sending piece of malicious code, which is often e-mailed, is called _____________. a. A worm b. A virus c. A Trojan d. Spam 10. What technique would an attacker do to force a wireless end device to disconnect from the network? a. Wireless scan b. Port scan c. OS fingerprinting d. RF jamming



42


11. What would a wireless hacker in his early teens most likely be doing? a. War flying b. War driving c. War walking d. War gaming 12. What does the term “spam” mean? a. Ham in a can b. Sending wanted e-mails c. Sending unwanted messages d. Sending junk snail mail



Chapter 3

The Legality of Computer Crime Throughout time, legislation has often been unable to keep up with technology growth. This is because of the rapid development of technology and the slow process of creating laws. It is important to understand what exactly constitutes a crime when any information security matter is involved. When discussing the legality of computer crime with respect to RF transmissions, a number of issues often arise. For example, in some states it is illegal to connect to an unauthorized network. With technology providing as much usability as possible, when does one cross the line? New wireless laptops are capable of connecting to insecure networks on their own. This means that merely walking down the street can be considered a crime. For example, the FCC can prosecute a person for eavesdropping, just for enabling a wireless sniffer and capturing packets from an unauthorized network. Think about this: when troubleshooting a wireless network in an office complex, a person might end up firing up wireless sniffers and finding out that the office suite next door has a wireless network. That person has just sniffed the airwaves and committed a crime subject to the FCC ruling. These examples reveal how different types of technology and legislation often are not finely tuned to each other, thus creating large gaps that allow criminals to exploit flaws and honest people to be criminalized by the same ones. One does not have to worry that using a

43



44


wireless sniffer will get one sent to jail. For example, it is illegal for people in Michigan to spit on the sidewalk. However, no one has been convicted of this crime in the past 50 years, although it is still listed as a law. This chapter shows the relevant legal matters affecting wireless networks. Some of the laws that were written for computer communications are not specific to wireless communications. Either way, most laws detail what constitutes a crime in relation to computers and network systems. The current chapter looks at the laws and acts that have created a good definition of what computer hacking is. Having a good understanding of that, this chapter goes on to discuss what crimes fall under these acts and laws. Just because a law says that breaking into a computer system is illegal, a process must be put in place to prevent every small computer crime from slowing down the court system. To fix this, certain requirements have been created to identify what types of crime are subject to these laws and acts, thus receiving federal aid and investigation.

3.1 Electronic Communications Privacy Act The Electronic Communications Privacy Act was enacted in 1986. This was to combat disclosure and interception. The main purpose is to protect the privacy of electronic communications. This law protects U.S. citizens from any inappropriate use of any intercepted electronic transmission. Electronic transmissions are defined by this act as “the transfers of signs, signals, writing, images, sound, data, or intelligence of any nature transmitted in whole or in part by wire, radio, electromagnetic, photo-electronic, or photo-optical systems that affect interstate or foreign commerce.” This act also requires government agencies to follow the correct procedure to obtain electronic communications from service providers. What this means to us has both positive and negative implications. The government must use due process and get a court order to tap any communications over an electronic device (e.g., a phone or e-mail). On the negative side of the act, it can allow an employer to listen and record an employee’s phone conversations, e-mail messages, and any other type of electronic communication taking place on the employer’s equipment. The last part of this statement is the most important. The employer must own and provide to the employee the electronic communication device. If the employer provides this device to the employee, then the employer can listen to the employee’s communications for a number of actions. Another key point is how and what employers can do with that information. The law has shown that if an employer wants to use any of the recorded communication for action against the employee, then the employer must first notify the employee that his or her communications are



The Legality of Computer Crime 45

being intercepted or recorded. An even more interesting point relates to free e-mail services on the Internet. These services can use a person’s email for a magnitude of purposes. They can sell your likes to related retailers, and they can market products directly to you based on your purchases. The main reasoning behind this is that the law states that the information residing on a service provider system is owned by the service provider even if it is your data. The bottom line to this statement is to be careful with any outsourced data endeavor; although contracts and privacy policies can protect you and your data, the law without these added documents does not provide any recourse from the misuse of data if it, in fact, resides on someone else’s system.

3.2 Computer Fraud and Abuse Act Enacted in 1996, the Computer Fraud and Abuse Act helped law enforcement fight the many computer crimes that were difficult to prosecute. This was because technology very rarely keeps up with legislation. This often creates many gaps that criminals exploit. This act allowed for the identification of what exactly constituted a federal computer crime in order to close these gaps. With a clear definition of what a computer crime is, this act has helped law enforcement agencies convict computer criminals who have evaded them in court thus far. The focus of this act is not to protect users, but rather to protect the federal government and its computer systems from any threat to national security. During the late 1980s and early 1990s, hackers were infiltrating government-controlled systems frequently. Without this federal law, prosecutors had trouble convicting these hackers of actually committing a crime. This is because of the lack of legislation dealing dir ectly with computer crimes and exactly what constitutes them as crimes. Now with the Computer Fraud and Abuse Act in place, this definition is clear, and the evasive computer hackers are now finding themselves in jail.

3.2.1 Patriot Act After September 11, 2001, the U.S. Congress passed the Patriot Act. This act amended Section 1030 of the Computer Fraud and Abuse Act and provided the government with expanded cooperation of service providers. They now were required to perform tracking of hacking, denial of service, and other computer crimes throughout their networks. When a computer crime is deemed a federal offense, the service provider is required to provide the records indicating where the offending party had initially



46


connected to the service provider’s network and what malicious actions they performed. By now, one must be asking what constitutes a federal computer crime, especially one that is subject to the Computer Fraud and Abuse Act. To constitute a federal investigation, there must be damage or loss that exceeds the average loss within that geographic area. This means that in a large, economically viable city such as New York, this could be as high as $100,000. In a small city out in a rural area, this amount could be as small as $2000 or $3000. An attack that only scans a network and does not compromise any data or create any denial of service would not qualify for a federal investigation.

3.3 State Computer Crime Issues During the late 1990s and into the turn of the 21st century, state governments had increasing problems dealing with small computer crime. This stemmed from a number of incidences where the computer crime was under the dollar limit required to constitute a federal investigation. On top of this comes another issue of pursuing suspects outside the state in which the victim lives. Very rarely is an item purchased online from a Web site owned and operated in the same state as the customer. This means that if someone were to fall victim to an online fraud incident, the state law enforcement agency would have no authority to apprehend a hacker or criminal outside the state in which that victim lived. This has created a massive problem in working together. Some might think that the easy answer would be to have them work together. This may seem true, although once one gets into the details of this, one starts to see the massive amount of small computer crimes compared to the small amount of law enforcement personnel dedicated to them. This has hampered the ability of state agencies to work together. This is because one state might pick a few cases to investigate and choose not investigate others. One state might not investigate a case that another state did investigate. This often happens and leads to the victim and the criminal being in different states. Once that is determined, the investigating state needs the help of the state that did not choose to investigate the case initially. As one can see, this is where the communication problem begins.



The Legality of Computer Crime 47

3.4 Chapter 3 Review Questions 1. What piece of legislation defines communication as “the transfers of signs, signals, writing, images, sound, data, or intelligence of any nature transmitted in whole or in part by wire, radio, electromagnetic, photo-electronic, or photo-optical systems that affects interstate or foreign commerce”? a. Patriot Act b. Electronic Communications Privacy Act c. Computer Fraud and Abuse Act d. Computer Crime and Abuse Act 2. What act detailed the use, disclosure, interception, and privacy of electronic commutations? a. Patriot Act b. Electronic Communications Privacy Act c. Computer Fraud and Abuse Act d. Computer Crime and Abuse Act 3. How much loss must a company have before it can get help from the FBI? a. $5000 b. $50,000 c. $500,000 d. It is based on location and the average loss for that location. 4. What did the Patriot Act amend relating to computer crime? Choose the best answer. a. Nothing; it was its own act. b. Electronic Communications Privacy Act c. Computer Fraud and Abuse Act d. Computer Crime and Abuse Act 5. When was the Electronic Communications Privacy Act created? a. 1999 b. 1988 c. 1985 d. 1986



48


6. When was the Computer Fraud and Abuse Act created? a. 1999 b. 2001 c. 1989 d. 1996 7. What act was created to help protect U.S. Government computers? a. Patriot Act b. Electronic Communications Privacy Act c. Computer Fraud and Abuse Act d. Computer Crime and Abuse Act 8. What act was created to protect U.S. networks? a. Patriot Act b. Electronic Communications Privacy Act c. Computer Fraud and Abuse Act d. Computer Crime and Abuse Act 9. Under the Patriot Act, who is required to log and track hacking attempts? a. The Department of Homeland Security b. The FBI c. The NSA d. The ISP 10. What regulation, act, or law was put into place due to the lack of clear terms regarding what constitutes a computer crime? a. Patriot Act b. Electronic Communications Privacy Act c. Computer Fraud and Abuse Act d. Computer Crime and Abuse Act 11. The FCC has a law that makes running a sniffer and receiving other people’s network traffic a crime. a. True b. False



Chapter 4

Wireless Physical Layer Technologies This chapter looks at wireless physical layer (Figure 4.1) technologies, which encompass the methods of radiating the wireless signal across the air. Many different techniques have been used over the years to send radio waves. When the IEEE got together to create each of the 802.11 standards, it decided on the method it felt would best provide the needed ability to send these radio waves as far and clean as possible. All the different methods used by the IEEE to send wireless signals today over the airwaves are based on a spread spectrum concept. Wireless networking utilizes this spread spectrum technology at its physical layer to communicate over the airwaves. Spread spectrum technology works by using different modulation techniques over a wider band of frequency than necessary. This makes spread spectrum signals look like noise, which makes them difficult to detect, intercept, or jam. The military has often favored spread spectrum technology compared to narrowband, which sends a signal on a very small frequency band at a very high power. Looking at Figure 4.2, one can see the difference between spread spectrum and narrowband. Military personnel used spread spectrum technology because they were able to transmit signals over such a wide frequency band that eavesdroppers could not distinguish it from radiant noise. Because of this, the military kept spread spectrum technology a secret until the 1980s when the FCC implemented some rules making it available to the public. The public release of the technology was to encourage research and investigation, leading to more advanced methods of transmission and securing 49 © 2006 by Taylor & Francis Group, LLC


50


Application Presentation Session Transport Network Data Link Physical

Figure 4.1 Wireless physical layer.

Narrow Band

Spread Spectrum

Power Levels

100 75 50 25

General Noise Level

0 1.0 GHz

1.5 GHZ

2.0 GHZ 1.0 GHz Frequency Range

1.5 GHZ

2.0 GHZ

Figure 4.2 Spread spectrum and narrowband.

wireless communications. This ended up leading to the 802.11 family of standards, which incorporates spread spectrum technologies. The FCC has always tried to conserve as much of the fr equency bandwidth as possible, only allowing what is needed for radio communications. Using narrowband transmission is how the FCC can conserve frequency and allow for numerous transmissions within a small radio frequency band. Narrowband communication requires a high power output to pass the signal beyond the inherent noise within any given air



Wireless Physical Layer Technologies 51

space. This inherent noise is often called the noise floor. As one might guess, this is particularly prone to interference and jamming because of the small amount of frequency being used. Using a wider band makes interference and the threat of jamming a reduced risk. This is why spread spectrum technology has been adopted by the military as a more reliable and robust means of communication. The history of spread spectrum technology is a very interesting one. It starts out with Hedy Lamarr, formerly Hedwig Eva Maria Kiesler, who was born in Vienna on November 9, 1913, to Jewish parents. In the late 1940s, she avoided the advancement of the German army by marrying an Austrian arms dealer. This man was one of the leading arms manufacturers in Europe. His name was Mandl and he operated a factory that supplied Hitler in his European campaigns. Hedy took on the role of hostess and entertainer for his guests. In doing so, she often overheard conversations from multiple arms dealers about the inherent problems with jamming torpedoes. She then thought that instead of using narrowband communication, which often guided those torpedoes at that time, they could, instead, use a technique that took the signal, spread it across a large frequency, and switched from one particular frequency to another inside the frequency range. Hedy, the woman who had learned about the latest in German and Austrian military technology at her husband’s plants, met composer George Antheil at a dinner party in 1940. This was where she shared what she knew about the design of remote-controlled torpedoes. Hedy believed the solution was to broadcast the weapons signal on rapidly shifting frequencies. She and Antheil developed a frequency-hopping system by which both the transmitting and receiving stations of a remote-controlled torpedo changed at intervals based on time. They received U.S. Patent 2,292,387 in August 1942, and the U.S. Navy put their research to limited use during World War II. What makes this even more interesting is who Hedy Lamarr became. She later became a movie star, appearing in many motion pictures. One scene that most defines her career is a ten-minute nude swimming scene in the movie “Ecstasy.” This movie made Hedy known as the first women to make a nude appearance on film. Hedy Lamar died on January 30, 2000, in her home in Orlando, Florida, at the age of 86. She had only just begun to benefit from her patent a mere three years earlier.

4.1 ISM Spectrum In 1985, the FCC modified Part 15 of the radio spectrum regulation. This modification authorized the use of wireless network products operating




Fre q

uen

cy

52

Time

Figure 4.3 Frequency hopping spread spectrum.

in the Industrial, Scientific, and Medical (ISM) bands. This free radio spectrum allowed the public, along with the commercial sector, to utilize and produce products that communicate in this spectrum. This opened up a large market for companies to sell wireless products transmitting on this spectrum. If one had a cordless phone in the late 1980s, one can see how this modification affected commerce. Operating on an FCC spectrum usually requires a license. In the ISM spectrum, this holds true to some degree. The FCC requires that it verify products to make sure they do not exceed the maximum power settings. Lucky for us, most products on the market do not operate with enough power to require a license. Any ISM spectrum product must meet certain requirements, such as operation under one watt for transmitter output power. The ISM spectrum is open for use by anyone as long as the devices operate within the guidelines set by the government. Many devices use the ISM spectrum for other than wireless networks. Microwaves and cordless phones are just a few devices that also share the 2.4 open ISM band. This means they can cause RF interference to any 2.4-GHz wireless networks in the area.

4.2 Frequency Hopping Spread Spectrum (FHSS) Frequency hopping spread spectrum (Figure 4.3; FH or FHSS) is a method for transmitting data on the physical layer of the 802.11 standard. This works by dividing the 83.5-MHz wide channels into a number of smaller frequency slots. Each country has a regulatory domain that specifies the allowed frequency slots. In the United States, the FCC has allotted for 79 channels between the 2.402- and 2.479-GHz band. When frequency-hopping communication takes place, the signal must transmit on a particular frequency slot for a given amount of time. The




time before the signal is required to hop is identified as the dwell time. Once this dwell time has expired, the frequency-hopping network is required to shift the communication into a different frequency slot. Once the dwell time is determined, there is a time delay in the electric circuitry to change the transmission frequencies from one slot to another. During this time, the radio is unable to transmit, creating a dead time (also known as the hop time). Frequency hopping proved resilient against narrowband interference. When a narrowband signal occupies a particular frequency band within the range of the frequency-hopping system, it will only interfere with the system for short time while it transmits on that particular frequency. What this means is that, during the course of communication, a narrowband signal might be transmitting over five of the 79 channels used, meaning interference points take place when the hop sequence inserts the data on one of the five channels being utilized for the other narrowband signal. With the available 79 channels and the hop sequence only landing on those five channels rarely, one can see the interference avoidance advantage of frequency-hopping systems. Frequency hopping applied to wireless networking was only able to yield 1- to 2-MB (megabyte) data rates. Frequency hopping was the dominant technology in the 802.11 standards. This was due to the placement of multiple frequency-hopping access points in a given area without the concern of them contending for the available radio frequencies. How wireless networking clients attached to these frequency-hopping networks was through the wireless beacons sent out by the access point. Chapter 5 discusses the beacon further. As applied to frequency hopping, these beacons contain the hop sequences. These hop sequences are what give the cards the ability to shift through the frequency slots and move to the next frequency slot that the access point was going to talk to. This is all based on the hop sequence that is predefined within the IEEE 802.11 standard. Table 4.1 shows the other allotted channels and frequency bands in countries outside the United States. When transmission takes place, the communication is divided among these frequency slots and a pseudorandom sequence is used to trigger the data to move into a dif ferent frequency slot within this 83-MHz band. These timing functions are called hop sequences, which are also regulated.

4.3 Direct Sequence Spread Spectrum (DSSS) The IEEE 802.11 as a physical layer transport method defi nes direct sequence spread spectrum (DSSS). With the proliferation of 802.11b, some changes were made to DSSS to match the 5.5- and 11-MB data rates



54


Table 4.1

Frequency Hopping World Channel Allocation

Country

Channel

Range (GHz)

Hop Size

United States

2 to 79

2.402–2.479

26

Canada

2 to 79

2.402–2.479

26

Britain

2 to 79

2.402–2.479

26

France

48 to 82

2.448–2.482

27

Spain

47 to 73

2.473–2.495

35

Japan

73 to 95

2.473–2.495

23

written into the new standard. This is sometimes referred to as high-rate wireless. The 802.11 standard has both FHSS and DSSS listed as a choice for physical layer transport. The IEEE had foreseen that FHSS would be a much more difficult protocol to adapt to a higher bandwidth scenario. With DSSS being an easier protocol to adapt to a higher bandwidth scenario, it was chosen. The reasoning behind originally putting FHSS in the 802.11 standard was for bandwidth and battery life. On an FH network, running multiple access points in the same location to provide bandwidth over the 2-MB standard was a big plus. Also noted by the IEEE was the increased battery life on FH devices compared to DS devices. These reasons are what keep FH in the 802.11 standard, unlike the 802.11b standard, which only chose a single spread spectrum. When the IEEE made 802.11b, it needed to also make it backwards capable or the existing market would not be so forthcoming to adopt this new technology. With higher data rates and backwards capability, 802.11b became a big hit and with it so did the underlying physical transport DSSS. This has made direct sequence (DS) one of the most widely used wireless LAN spread spectrum techniques today. Direct sequence uses two of the modulation techniques we looked at in chapter one, BPSK and QPSK. Direct sequence is a method of transmitting data across the air on a 22-MHz-wide frequency range. This is done to prevent narrowband noise or interference. This is performed by sending the signal across a large bandwidth of frequency and eliminating any small narrowband noise. When this narrowband interference appears, which is often in short small spikes, it is easily distinguished from a slow-rising or sloping wave like the ones in DSSS. To have these larger, less interferenceprone waves, a large frequency bandwidth is needed. This large bandwidth eats up the small amount of frequency allocated in the ISM spectrum. Governmental bodies regulate allowed channels that can exist within this 2.4-GHz range. Looking at direct sequence channels, one notes that there are 14 available channels. Table 4.2 illustrates that each of these © 2006 by Taylor & Francis Group, LLC


Wireless Physical Layer Technologies

Table 4.2

55

Direct Sequence Spread Spectrum World Channel Allocation

Frequency

Channel

Americas

Middle East

Asia

Europe

Japan

Israel

2.412

1

X

X

X

X

X

2.417

2

X

X

X

X

X

2.422

3

X

X

X

X

X

X

2.427

4

X

X

X

X

X

X

2.432

5

X

X

X

X

X

X

2.437

6

X

X

X

X

X

X

2.442

7

X

X

X

X

X

X

2.447

8

X

X

X

X

X

X

2.452

9

X

X

X

X

X

X

2.457

10

X

X

X

X

X

2.462

11

X

X

X

X

X

2.467

12

X

X

X

X

2.472

13

X

X

X

X

2.477

14

channels starts 5 MHz from each other and each has a bandwidth of 22 MHz. This makes these channels often interfere with each other. The proper designing of wireless networks must take into account this inherent overlap and assign proper channel selection. Looking at the channels available in the United States, one notes that there are three nonoverlapping channels: (1) channel 1, which operates at 2.401 to 2.423 GHz; (2) channel 6, which operates at 2.426 to 2.448 GHz, and (3) channel 11, which operates between 2.451 and 2.473 GHz. Placing all these channels on a linear path gives one a better feel for their relation and inherent interference between the selectable channels. A full list of all the channels and their frequencies for North America is listed in Table 4.2. Looking at Table 4.2, channels 1, 6, and 11 do not overlap each other. The exclusive uses of these channels are recommended in a best practice design. Understanding what causes overlapping channels is a big part of being a successful RF engineer. To understand channel overlap, take a look at the following example. Case 1: you have stuck with the design guidelines and have implemented on channels 1, 6, and 11. Now that you are done, there are some parts of the facility where you have channel 6 overlapping with itself. This creates a large, layer two domain, such as daisy chaining two hubs together. This would make both areas of coverage © 2006 by Taylor & Francis Group, LLC


56


Direct Sequence Spread Spectrum Channel Allocation

Frequency 2.400 2.405 2.410 2.415 2.420 2.425 2.430 2.435 2.440 2.445 2.450 2.455 2.460 2.465 2.470 2.475 2.480 Channel Center Frequency

1 2 3 4 5 6 7 8 9 10 11 2.412 2.417 2.422 2.427 2.432 2.437 2.442 2.447 2.452 2.457 2.462 GHz GHz GHz GHz GHz GHz GHz GHz GHz GHz GHz Copyright Aaron Earle 1/20/2005

Figure 4.4 Direct sequence spread spectrum channel overlap detail.

the same broadcast domain. This is difficult to avoid and appears multiple times. To overcome this, one can try moving to a four-channel design. This may increase the network throughput, although it is a trade-off between large broadcast domains and slight channel interference. Performing throughput testing using both methods would allow an engineer to adapt the best method to deal with the limited nonoverlapping space. To recap, if one is in a situation where one might need to use a fourchannel design, perform some throughput testing to see if channel interface or a large broadcast domain produces better throughput in the particular situation. Various countries limit the use of the available channels. Looking at Table 4.2, one can see each of the countries and their allowed wireless DSSS channels. In the United States, only the use of channels 1 through 11 are allowed. The United Kingdom can use channels 1 through 13. Finally, Japan uses all 14 channels. Be aware this can complicate matters when designing international wireless local area networks (LANs). This can also be seen when you deal with end devices that travel between these countries. When this happens, one can take two approaches: (1) use 802.11d to dynamically adjust client devices when connecting to an access point in different countries or (2) just design setting the common channels across all access points globally. This can be done by never using channels 12 through 14 in Europe or Japan. © 2006 by Taylor & Francis Group, LLC



4.4 Orthogonal Frequency Division Multiplexing (OFDM) In July 1998, the IEEE selected Orthogonal Frequency Division Multiplexing (OFDM) as the basis for the 802.11a standard in the United States. OFDM is a wireless technology patented in 1970. As seen in Chapter 1, OFDM works with four modulations: BPSK, QPSK, 16-QAM, and 64-QAM. OFDM works using the allocated frequency band and dividing it into a number of low-frequency sub-carriers. In 802.11a, the 20-MHz channel is divided into 52 such sub-carriers that are 312.5 kHz wide running across eight standard channels. These sub-carriers are orthogonal, meaning every sub-carrier can be separated out at the receiver without interference from other sub-carriers. This is possible because the mathematical property of orthogonal waveforms ensures that the product of any two sub-carriers is zero. This also allows the receiving end to reconstruct damaged portions of the signal. OFDM technology can be seen in broadcast systems such as Asymmetric Digital Subscriber Line (ADSL), Digital Audio Broadcasting (DAB), Digital Video Broadcasting Terrestrial (DVB-T), 802.11a, and 802.11g.

4.5 Chapter 4 Review Questions 1. An 802.11a radio uses what technique to transmit its signal? a. FHSS b. DSSS c. OFDM d. OFSS 2. The 802.11b standard does not use DSSS. a. True b. False 3. Which standard listed below supports Frequency Hopping Spread Spectrum? a. 802.11b b. 802.11g c. 802.11a d. 802.11



58


4. Within North America, what number of channels are considered nonoverlapping? a. 1 b. 3 c. 14 d. 11 5. The 802.11g standard supports the use of orthogonal frequency division multiplexing (OFDM) in which radio frequency band? a. 2.4 MHz b. 5.4 MHz c. 5.8 MHz d. 5.0 MHz 6. Dwell time is a function of which of the following physical layer techniques? a. DSSS b. FHSS c. OFDM d. PHY 7. DSSS is a type of what? a. Spread surface b. Spread spectrum c. Modulation type d. Network identifier 8. How many channels of DSSS are in use in the United States? a. 1 b. 14 c. 12 d. 11 9. How many channels are used in an ESTI-governed location? a. 1 b. 13 c. 11 d. 14



Wireless Physical Layer Technologies

59

10. The 2.4 frequency that 802.11b operates in is called __________. a. DSSS b. OFMD c. UNII s. ISM 11. The 802.11a standard is associated with which two terms? a. ISM b. UNII c. 2.4 GHz d. 5.15–5.30 GHz 12. How much of the 2.4-GHz ISM band is used when a single access point 802.11b wireless network is enabled? a. 11 MHz b. 22 MHz c. 1 MHz d. 20 MHz 13. What is the total amount of wireless bandwidth inside the 2.4-GHz ISM band? a. 70 MHz b. 11 Mbps c. 54 Meg d. 83.5 MHz 14. What is the amount of time a channel takes to switch from one channel to another channel on an FH system called? a. Hop time b. Dwell sequence c. Hopping time d. Dwell time 15. You are designing an RF network and have channel 6 used. What other channel would you select to use next to channel 6? a. 4 b. 3 c. 1 d. 9



Chapter 5

Wireless Management Frames This chapter focuses on the data-link layer (Figure 5.1) of the Open System Interconnect (OSI) model. This is where wireless frames are defined. The data-link layer is the second layer of the OSI model; it takes information and divides it up into frames. These frames are forwarded to the first part of the OSI model, the physical layer, to be transmitted out of the wireless network’s antenna. While Chapter 4 focused on wirelesses physical layer technologies and how they work, this chapter explores wireless frames. When most people think of wireless Ethernet, they think it is very similar to wired Ethernet. This is not the case; Ethernet was created upon 802 just like wireless, although Ethernet is defined as 802.3 and wireless is defined as 802.11. The difference here is how the MAC layer information is handled. In Ethernet, the maximum frame size is 1518 bytes; wireless has a maximum frame size of 2346 bytes. There are some major differences between wireless Ethernet and normal wired Ethernet. This chapter sheds some light on this as it discusses the mechanisms used for wireless frame management. Some of these mechanisms are very similar to wired Ethernet and some are very different. There are three main types of frames used inside the 802.11b/a/g wireless standards. These frames make up all wireless communications. They are control frames, management frames, and data frames. Control frames control the data flow; management frames allow users on and off the network; and most importantly, data frames move the precious data. Looking at Figure 5.2, one can see a general 802.11/b/g/a frame and that many parts make up just a single frame. Looking more closely at the 61



62


Application Presentation Session Transport Network Data Link Physical

Figure 5.1 Wireless data-link layer.

frame’s control section, one sees that this 2-byte portion contains a large amount of identification information. Looking at each part, one can see exactly what its function is. Protocol Version. This is a field to identify the frame as a wireless frame. Currently, there is only one setting for this field. Type. This part of the frames identifies the frame as a control frame, a management frame, or a data frame. Subtype. This portion of the frame further identifies itself as one of the various types of control, management, or data frames. To DS. This field indicates if the frame is destined for a Distribution System, which identifies a wireless network. From DS. This field indicates if the frame is sourced from a Distribution System. More Fragments. This field is used to indicate that there are more frames arriving. This means other fragments or portions of this frame will be following this frame. Retry. This field informs the receiver that the frame has been already sent and that it was informed to retry the transmission. Power Management. This field is used to identify if a device is in the power save mode. Access points do not go into power save mode and always send or set this value to zero. Clients, on the other hand, do go into power save mode and use this frame indicator.


2 Bytes

Subtype

To DS

From DS

2 Bytes

2 Bytes

4 Bytes

1 Byte

1 Byte

Address 1 6 Bytes

Figure 5.2 Wireless frame detail.

More Fragments 1 Byte

Retry 1 Byte

Power Management 1 Byte

More Data

WEP

Order

1 Byte

1 Byte

1 Byte

Address 2

Address 3

Sequence Control

Address 4

Frame Body

Frame Check Sequence

6 Bytes

6 Bytes

2 Bytes

6 Bytes

0–2312 Bytes

4 Bytes

Wireless Management Frames

2 Bytes

Frame Control


© 2006 by Taylor & Francis Group, LLC Frame Duration Control

Frame Control

63


64


More Data. This field is used to inform clients that the access point has received multiple data frames for the receiver and has started to fill its buffer with them. WEP. This field is used to indicate if WEP encryption is or is not being used. Order. This is a portion of the frame to help with Quality of Service (QoS).

5.1 Beacon Access points identifying themselves and their configuration settings often send out beacon frames. Clients can send out beacons only when in ad hoc mode. These beacons contain the access points’ supported data rates, the channel that it is talking on, and if it supports WEP or not. Depending on the settings, it might also broadcast the SSID. The beacon frames are used for wireless clients to understand who can provide them service in a given area. A beacon frame is like an access point screaming out how it is configured so that clients can hear this.

5.2 Probe Request Probe requests are similar to that of a beacon in that they are both used for identification and configuration information. Unlike the beacon, which transmits the identification of itself out over the air waves to any device in that area, the probe is used by the wireless client to locate an access point with the similar identification settings. If the client’s probe has the correct SSID, then the two devices determine if their settings allow them to start the authentication process. This probe request contains two main parts: (1) the SSID and (2) the supported data rates. Probe requests are often used to find access points that are not sending beacons. Access points can suppress their beaconing, so a mechanism to find them must also exist. This is where the probe request fits in.

5.3 Probe Response An access point that hears a probe request will respond. It will only respond when the SSID from the request matches what the access point is configured to use. This probe response packet will identify the configuration settings of the access point. The probe response packet will have the following settings: data rates, SSID, and channel information. This is



Wireless Management Frames 65

similar to access point beacon, although it will only be sent if a client places a correct probe request that matches identification information with the access point.

5.4 Authentication This section identifies the means of authenticating to a wireless network. How the authentication is accomplished depends on the security settings and the level of enabled security. Authentication must take place before attempting to connect to a wireless network. This frame has a number of identification mechanisms to identify and authorize a wireless device. Looking at the frame in detail, an authentication frame has four main parts, which are unique to it: 1. Authentication algorithm: a portion of the frame identifying what security settings it will be using. 2. Authentication transaction number: identifies which frames are parts of which authentication transaction. This is used when an access point is serving multiple authentication attempts simultaneously. 3. The status code: used to identify why a failure or success occurred. 4. Challenge text: a means of security by which a cleartext packet is sent for security purposes.

5.5 Association Request Once a wireless device has successfully authenticated, it can then start the next step of joining the network, called association. To associate, the wireless client must again send its configuration information to the access point. The access point looks at this information and makes a decision of whether or not to join the device to the network. This decision is not based on authentication because that has already taken place before this step; it only verifies that the settings between the two are agreed upon by both parties. These settings are SSID, channel, and power.

5.6 Association Response This frame, which is generated by the access point, identifies to the wireless device that it has been connected to the network. Association response frames are closely related to association request frames. The



66


only difference between the two is a small identification attribute in the frame. Once the client gets this frame, it knows that they have successfully connected to the network and are ready to start utilizing network resources.

5.7 Disassociation and De-Authentication These two frames are the same, with the only difference identifiable through the reason code. This code is embedded in the body of the frame and is used to identify why a client was removed from a wireless network. The reason code is a small 2-byte portion of the frame. This reason code is capable of identifying more than 50,000 reasons; luckily, there are only ten currently specified reasons for this field. The reasons used inside this code are laid out from least significant to most significant, running from 0 bytes as the least significant to 65,535 bytes as the most significant. Both of these frames are used to remove a wireless device from the network. The main differences between these two frames are the reason for why they removed a device and how or what piece of equipment made the determination to remove the wireless device. The Disassociation message is used by the access points to remove clients that have not talked for a particular amount of time. This allows the access point to clean up after itself and remove old clients that have disconnected without notice. The De-Authentication frame is sent by an administrator is remove a wireless device from the network.

5.8 CSMA/CA Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) is the technology that operates at the data-link layer. This second layer of the OSI model responsible for moving frames across its known view of the world or its network segment. This is similar to Ethernet’s Carrier Sense Multiple Access with Collision Detection CSMA/CD, which is a second OSI layer protocol to move frames in an Ethernet environment. The difference between the two is more than what media the frames are moving across. With wireless networking having bandwidth at such a premium, a scheme with an acceptance for bandwidth loss was not exactly the best approach. On the wire, losing a small amount of bandwidth to speed up the movement of network frames was considered acceptable. This was where the last two words in the CSMA/CD acronym stand for “collision detection.” In CSMA/CD, it detects that a collision took place and resends the frame. On a wireless network losing bandwidth along




67

with time to detect that a collision took place was not acceptable, so another mechanism (called collision avoidance) was utilized. The process of how CSMA/CA works is detailed below, including each of the four critical steps involved in a transmission and how they work together to move frames across a wireless network.

5.8.1 RTS The first step in this process is the initial request. This is called the Request to Send or, for short, RTS. The client computer initiates the RTS when it needs to send data to the access point or through the access point to other network resources. As part of the initial request, the client determines the amount of time he or she needs to complete the transmission. This is set with the NAV bit. The access point will use this to make a decision on how long the client can talk before it has to ask permission again. Looking at Figure 5.3, one can see that this 20-byte frame has five different parts. First is the frame control section, which is a field identifying the frame as RTS. Next is the duration portion, which signals to the access point the amount of time that the client wishes to have the airwaves for to send information. Next is a section called the receiver address; this address is the MAC of the sending station. Next is the transmitting address section, which is also composed of a MAC address. The final portion consists of the frame check sequence. This is the wireless version of an Ethernet checksum. With the wireless airwaves being less reliable media than standard Ethernet, a different technique was used to send a checksum. This technique works by sending a checksum for the previous frame.

Frame Duration Control 2 Bytes

Receiver Address

2 Bytes

Transmitter Address Frame Check Sequence

6 Bytes

6 Bytes

4 Bytes

Figure 5.3 Request to send frame detail.

Frame Control

Duration

Receiver Address


2 Bytes

2 Bytes

6 Bytes

4 Bytes

Figure 5.4 Clear to send frame detail.



68


5.8.2 CTS Next, one sees the request as it is returned by the access point with a clear to send frame (CTS). This frame indicates how the long the client can talk for before having to close the CSMA/CA process again. This CTS frame consists of segments to the RTS frame. Looking at Figure 5.4, one can see the frame control segment, which announces its existence as a CTS frame. Next is a duration segment, which makes up the amount of time the wireless device has to transmit data. Next is the receiver’s MAC address, which is copied from the previous frame’s transmitting address, and finally the frame check sequence.

5.8.3 DATA This section does not deal with any management per se, however, the data section is needed to fully understand what sequence of events takes place when a wireless device wishes to transfer data. Looking at how this plays a role in the entire data transmitting sequence, one can see that the data frame is sent and subsequent data frames continue to be sent until the time allotted by the CTS frame has been used. Once this happens, the data is finished transmitting until a new CTS frame is received with a new period for which the data frame process can resume. To get this new CTS frame, the entire process must be restarted with an RTS. As one can see from Figure 5.5, this frame looks a little different because it is actually a data frame and not a management frame like the RTS and CTS frames.

5.8.4 ACK Acknowledgment frames are sent by many different networking protocols to identify to the receiver that the data sent was received and was not corrupted en route. This acknowledgment frame is made up of similar sections such as the Request to Send and Clear to Send frames. Looking at Figure 5.6, one can see that, like the other frames, the acknowledgment has a frame control section that identifies it as an acknowledgment frame. It has a duration portion that is used to inform the access point if this acknowledgment is the last one in the current data transfer process. Finally, there is the receiver address section, which is copied from the transmitter’s MAC.

5.9 Fragmentation Fragmentation is common on wireless networks due to inherent interference. Its function is not what many people think. Fragmentation occurs


Destination Address

BSSID

Source Address

Sequence Control

Payload


2 Bytes

2 Bytes

6 Bytes

6 Bytes

6 Bytes

2 Bytes

0–2312 Bytes

4 Bytes

Figure 5.5 DATA frame detail.


Duration


© 2006 by Taylor & Francis Group, LLC Frame Control

69


70


Frame control

Duration

Receiver address

Frame check sequence

2 Bytes

2 Bytes

6 Bytes

4 Bytes

Figure 5.6 Acknowledgment frame detail.

to enable the wireless media to be able to send more data. This is done with the use of smaller transmission. With wireless being such an unstable medium, sending large frames and then, in the event of corruption, having to resend them takes more time than sending and resending smaller frames. Because of this, fragmentation is often performed to speed up the transfer of frames. It is purposely applied to break up the transmission into small pieces that can be resent very quickly. This is important because the wireless medium is only allocated to a single user for a set amount of time and then everyone must contend to use it. This fragmentation can cause problems as well; fragmentation creates overhead and the processing time requires splitting the frames and reassembling them on the other end. Setting fragmentation can be a tricky option. One needs to make sure it is set between having it over-fragmented and causing processing issues, or having it under-fragmented and causing retransmit errors. Usually, access point manufacturers have tweaked this number to its most optimal setting, although in some cases it may need on-site adjustment.

5.10 Distributed Coordination Function The distributed coordination function (DCF) is used to poll the media or, in this case, the wireless to see if someone else is using it. As seen above, before one can send any frames, one needs to make sure no one is using the network. The DCF performs this by examining the entire MAC header of frames that are sent. When another machine wants to use the network, it requests access, stating how long it needs the network for its own use. When it does this, its frames are tagged with this indication, called the network allocation value (NAV).

5.11 Point Coordination Function The point coordination function (PCF) has the same goal as that of the DCF, although it handles it differently. In the PCF, a device tells the access point that it can respond to polls. This means that PCF can only work



Wireless Management Frames 71

within a wireless network controlled by an access point. Once the access point is aware that the device can support polls, it will poll the entire wireless portion, asking if clients need to send any frames. This provides the device that needs to transfer information with the ability to do so. The PCF must function in unison with the DCF. This is not the other way around; the DCF can function without the PCF.

5.12 Interframe Spacing To perform management functions and allow for time synchronization, a preset interval where nothing communicates is required. This is termed “interframe spacing.” This value is set into all the devices as part of each 802.11 standard. This value allows for access point management and the prioritization of network traffic. There are four types of interframe spacing: 1. 2. 3. 4.

Short interframe space (SIFS) Point coordination function interframe space (PIFS) Distributed coordination function interframe space (DIFS) Extend interframe spacing (EIFS)

5.13 Service Set Identifier (SSID) When looking at the Service Set Identifier (SSID), one sees that there was never any intention of having it perform any type of security measure. The SSID’s main purpose is for network identification, as the name states. When a client end device connects to the network, it must have an identification setting to allow it to know what network to connect to and operate on. When the wireless standards were created, the IEEE had the foresight to realize that there may be more than one wireless network within range. This led to the creation of the SSID as a means to differentiate one wireless network from another. Today, with the massive amount of wireless networks, this has become a necessity. This is what the SSID is for: network identification. The SSID can also be used to create multiple virtual wireless networks. This is very similar to VLANs, which are used in the wired world. Having multiple wireless networks does mean that everyone is still sharing the same air space, although they are on their own wired subnet. This is often used to accommodate guests and allow for different security levels. This can be seen in a situation where some older devices could not support the advanced security standards. In this situation, two wireless VLANs



72


could be created: (1) one with the advanced security open to go anywhere in the wired network and (2) the other one that has support for the older weaker method of the security. The weaker method of security would also have some other wired security methods such as ACL or IDS, or some other added security mechanism to balance out the higher wireless security risk it brings.

5.14 Chapter 5 Review Questions 1. What are the three main wireless frame types? a. Control frames b. Management frames c. Data frames d. Session frames 2. When a wireless network is using an access point, which device sends the beacons? a. Access point b. Wireless laptop c. Authentication server d. User 3. Which layer 2 protocol does 802.11b use? a. 1CSMA/CD b. RTS/CTS c. DSSS d. CSMA/CA 4. What is the difference between a Disassociation and De-authentication frame? a. Frame TTL b. Frame size c. Reason code d. Protocol field 5. CSMA/CA is a _____________ process. a. One-step b. Two-step c. Three-step d. Four-step




73

6. To perform management functions and allow for time synchronization, a preset interval where nothing communicates is required. What term does this describe? a. DCF b. PCF c. CTS d. Interframe spacing 7. Which step takes place first: authentication or association? a. Authentication b. Association 8. CTS/RTS is part of what? a. CSMA/CA b. CSMA/CD c. WECA d. IEEE 9. Which layer 2 protocol does Ethernet use? a. CSMA/CD b. RTS/CTS 3. DSSS d. CSMA/CA 10. What is the SSID used for? a. Security b. Identification c. Information exchange d. Network management 11. The distributed coordination function (DCF) is used to _________ . a. Poll the media for network identification b. Poll the media for security c. Poll the media for network availability d. Poll the network for any changes



Chapter 6

Wireless Local and Personal Area Networks What are wireless local area networks (WLANs), and how do they work? This chapter explores what makes up wireless networks, how they are set up, and what kind of hardware is involved. It then looks at how the market has adapted to creating many wireless standards for interoperability between multiple vendors. This chapter introduces the main wireless standards and describes how they work and what differentiates them. At first, wireless networks were low data rate, proprietary installations that were used only for the most needed mobile applications. This was common in warehousing and other inventory management solutions. Today’s wireless networks have gained momentum in a number of vertical markets such as healthcare, education, retail, manufacturing, warehousing, and more. This growth is directly related to the fact that wireless networks give users the ability to remove constraining wires. This freedom from wires allows users to work anywhere there is a network signal present. Wireless networks bring massive gains — not only in productivity, but also from reduced cabling and fast client relocation. All these advantages have made wireless networks pop up all over the corporate landscape and subsequently created this fast-growing industry. Flexibility is a major reason that wireless networks have become so popular. Just looking at historical buildings gives us an example of this. Once a building is deemed historical, running wires through it can quickly



76


become an unacceptable option. With wireless networks, no wires are necessary; a user just has to plug into an access point and he is set to go. Without having to drill holes for wires, these historical buildings can keep their old-world look and feel. Another way the flexibility of wireless networks is useful is in areas or buildings not owned by the occupant. In this case, holes cannot be drilled into the walls to install wire runs. Wireless allows one to set up the access point and connects all the needed information systems via a wireless connection. Disaster recovery is another area where the flexibility of wireless plays a key role. When major damage impedes the ability to hang cables, using wireless can help keep a workforce connected. This can be seen in events such as the tornado that hit the General Motors Oklahoma Assembly Plant in 2003. To help speed up the construction effort, wireless bridges were put in connecting temporary trailers to the GM infrastructure. This gave workers the ability to be connected to the communication tools they needed, thus allowing them to rebuild the plant faster. Recently, the phone companies have also jumped on the bandwagon, releasing pay-for-use wireless networks located in high-traffic areas. These are called hot spots and can be located in airports, hotels, coffee shops, and many other prime places throughout the world. Just connect and input a credit card or subscription information, then surf the Web, e-mail, or VPN into your corporate network. With market research companies like Gartner reporting hot spot availability going from a little over 1000 in 2001 to 130,000 in 2004, we are likely to see hot spots covering a large portion of heavily populated cities within a few years.

6.1 Ad Hoc Mode Wireless can operate in two main modes. The first one is called Ad Hoc mode, or Independent Basic Service Set (IBSS). This is a peer-to-peer wireless network. This means that it does not have an access point controlling the conversation. This mode is popular for small deployments where the number of users is less than five. In this method, the access point that usually manages the conversions is gone and the clients send beacons to each other instead. From Chapter 5 we learned that this is the only case in which a client would send a beacon frame. These beacons contain a timer synchronization function (TSF) to ensure that the timing is correct. This function is usually handled in the access point. Figure 6.1 shows what an Ad Hoc wireless network looks like. As one can see, there is no access point. This method of wireless was created primarily not to circumvent access points, but rather to allow two or more computers to



Wireless Local and Personal Area Networks

77

Figure 6.1 Ad Hoc or Independent Basic Service Set (IBSS) diagram.

Figure 6.2 Infrastructure Mode or Extended Basic Service Set (EBSS) diagram.

share each other’s resources when together. This helps traveling teams and workers exchange ideas and information more easily in locations where networks do not exist.

6.2 Infrastructure Mode The more commonly adopted mode for wireless is the Infrastructure mode, which is called Extended Basic Service Set (EBSS). This is the main type of wireless network. In an EBSS, an access point controls all traffic. Setting up a wireless network in this category requires a piece of networking equipment referred to as an access point. This access point is where the Ethernet data is converted into a wireless signal that is then transferred out through the access point’s antenna. To hear and understand this signal, a wireless network interface card is needed. This card has a small antenna inside it and can hear the wireless signal and transfer it to the computer. This mode is seen in Figure 6.2. This allows a group of people to connect to the network in any area covered by the access point’s signal range.



78


Wireless Bridge

Figure 6.3 Wireless bridge.

6.3 Bridging Wireless networks can also extend existing wired networks (see Figure 6.3). This is most often found in bridging applications of wireless. Bridging can be used for a variety of reasons, including connecting campus buildings together, connecting nearby businesses together, or supplying last mile service as an ISP. When wanting to connect two buildings together, instead of paying the high costs of leased lines, one can build a wireless bridge. The ability to build a bridge requires that the buildings be no more than ten miles apart. However, technology and cost can overcome this distance limitation. For example, satellites can be used to bridge any two buildings in the world. Usually, for normal application, ten miles is acceptable. The bridge can also be applied for companies in the same vicinity that do business with each other. These businesses might need to connect together their inventory systems. In the case of the last mile, the ISP will install a wireless bridge on the roof of a home or business and connect it to ISP’s local office. This provides Internet services via wireless and saves the ISP from incurring the cost involved in running wires to the home or business.

6.4 Repeater A repeater (Figure 6.4) is an access point that supports a feature allowing it to use wireless instead of wired Ethernet to connect to the backbone. When this is done, the access point that is acting as a repeater has no physical connection to the network. It is connected to the wired network




79

2900 AP wired to switch

Figure 6.4 Wireless repeater.

through a wireless connection to another access point that is wired into the network. This is common for large places that do not have the ability to run cabling and are too large to cover with a single access point. The downside to using a repeater is that all wireless bandwidth for the repeating access points and the one serving the connection to the wire are all the same. This means that everywhere there is a repeater, all clients connected to the repeater, and the other access points are contending for the same amount of limited bandwidth.

6.5 Mesh Wireless Networks Some ISPs, manufacturers, and a number of municipalities have started to use wireless mesh networks. Just like the Internet where routers have many paths to take to get to point A from point B, wireless mesh networks use similar routing protocols to create one highly available network. As seen with hot spots, these networks provide similar functionality on a much larger scale. This form of wireless has been primarily used in citywide or other very large wireless networks. Its main advantage is that not every access point must be hard-wired into the network. As with repeaters, some access points can function in this manner while others have direct connections into the network. Many cities have started using these types of networks to provide wireless Internet access similar to garbage disposal or other city services covered under taxation. These new types of networks have started to create new access points that make routing decisions like



80


a normal network router as well as providing wireless access. These new network types have also created the need for wireless routing protocols.

6.6 Local Area Networking Standards Once the value and cost reduction of wireless networking became widely known, many companies wanted to start using a wireless network. This created some problems at first because wireless networks were manufacturer specific; and if that manufacturer went out of business, so did the wireless system. To make things worse, all the end devices needed to be changed if the wireless network was upgraded or moved from one manufacturer to another. This was because all wireless networks were manufacturer specific and proprietary. This made the industry push the IEEE to make some wireless standards and help facilitate the growth of wireless with common standards that allowed various manufacturer cards to work with various manufacturer wireless networks.

6.6.1 802.11 In 1997, the IEEE standardized the 802.11 standard. This was the first WLAN standard accepted by multiple vendors as a true industry standard. The IEEE defines how different types of media carry data. The scope of the IEEE is broad; it writes standards for Ethernet, fiber optics, Token Ring, wireless of course, and many others. The 802.11 standard defines the Media Access Control (MAC) and physical (PHY) layers for a wireless network. Above these levels, there is the network layer, which has IP as a well-defined standard. Why none of these standards accounts for IP is that IP is non-caring to what media it is passing over. These networks operate on two physical layers: (1) direct sequence spread spectrum (DSSS) and (2) frequency hopping spread spectrum (FHSS). Each uses a different method of transmitting wireless signals across the airwaves. DSSS uses a wide, single, statically defined channel that is preset in the access point. On FHSS or FH, the access point and the client negotiate a hop sequence, which is used to allow the signal to switch between small slices of frequency in the 2.4-GHz range that wireless 802.11 has defined as usable. The MAC layer has been standardized to help contend with the interference and excessive loss of frames compared to Ethernet. The 802.11 standard has a maximum data throughput limit of 2 MB. This hindered desktop use of wireless networks. With 100-MB networks on the market and 1-GB networks less than a year away, desktop users were not willing




81

to work with a fraction of their old 10-MB networks, which they fought so hard to have upgraded. This limited 802.11 wireless networks and made mainstream wireless network use a pipedream. These networks were widely deployed for inventory tracking and r etail point-of-sale machines. The low data rate was not a problem for small hand-held scanners and other applications. These applications grew because they now were mobile and could better improve worker productivity.

6.6.2 802.11a In 1999, the IEEE group successfully standardized the 802.11a standard. The 802.11a and 802.11b standards were created at the same time. What delayed the arrival of 802.11a products versus 802.11b was that 802.11b products were already shipping as the IEEE standard was being finalized. Many 802.11b products were released well before 802.11a. This happened because many companies already had performed the research and development involved with 802.11b products and not 802.11a. Once the 802.11b products were complete and being sold, the research and development focused on 802.11a products. The 802.11a standard allows for 54 MB of data to be used on the airwaves. The physical layer technology Orthogonal Frequency Division Multiplexing (OFDM) is used to transfer the data into radio waves. The next technology, which controls the flow and management of the data, is the Carrier Sense Multiple Access with Collision A voidance (or CSMA/CA). The 802.11a standard uses 52 sub-carriers, four of which are used for monitoring, path sifts, and inter-carrier interference. Depending on where in the allowable frequency range one places the 802.11a device or radio, one must be aware of government requirements before transmitting. There are three sets of regulations with respect to power and antenna use with 802.11a. All three are called the Unlicensed National Information Infrastructure (UNII) spectrum and are identified as UNII 1, UNII 2, and UNII 3. The UNII 1 specification calls for indoor use only and operates in the 5.15- to 5.25-GHz range. It uses four of the available eight 802.11a channels. When using an 802.11a channel in this frequency, one must only provide it a maximum power of 40 milliwatts (mW). One big thing to note about UNII 1 is the requirement for fixed antennas. This means antenna options are not allowed unless they are affixed to the radio by the manufacturer. Looking at UNII 2, one sees that it has four channels and operates in the 5.25- to 5.35-GHz frequency range. There is a limit on its transmitting power, which is restricted to 200 mW. The UNII 2 spec allows for the use



82


of external antennas. Most 802.11a equipment today uses a combination of UNII 1 and UNII 2; when using these two frequencies together, one must follow the UNII 1 rule about fixed antennas. New changes have taken place in the FCC regulations allowing more channels and different rules. In February 2004 the FCC made changes that allow for the use of 11 new channels. This will give 802.11a designers 23 available channels. Another primary carrier of the frequency is a type of radar. What this means is there is another carrier or user operating in this frequency range. In this case this primary user is radar; because of this a mechanism must exist to limit the amount of interference. To help keep these two signals inference free, the IEEE created a standard called 802.11h. This standard is explained within this chapter. This standard allows for the ability to sense the usage of radar in a given area and change the access point’s signal away from those channels. The FCC mandate that any 802.11a users that want to use the new 11 channels must be 802.11h compliant. The UNII 3 frequency range jumps a little ways down; unlike the consecutive UNII 1 and UNNI 2, UNII 3 is not consecutive. The UNII 3 range is 5.725 to 5.825 GHz; this is interesting because UNII 3 somewhat overlaps the ISM upper frequency. This does not affect any 802.11 products in the ISM range because they run in the 2.4 range of the ISM. Some of the new 5.8-GHz cordless phones might have similar problems that 802.11 faced, although 5.8 GHz does not interfere with UNII 1 or UNII 2, which are the most prevalent 802.11a wireless transmission methods. The 802.11a standard has one of the smallest ranges in the 802.11 family. Looking at the maximum range for data rates above the 11-Meg 802.11 standards, one sees that it is only 90 feet. With 802.11a, one has the ability to place multiple access points in the same area, overlapping their coverage cells to provide higher bandwidth. This allows one to have more users per area, sharing a larger amount of bandwidth. This is possible because 802.11a allows for greater channel usage without running into overlapping channel issues that plague DSSS systems.

6.6.3 802.11b The IEEE created the 802.11b standard at the same time (1999) as the 802.11a standard. This standard arose from the need to have greater bandwidth than what was available in the then-current wireless 802.11 standard. This new standard was required not only to have higher bandwidth, but also to be backwards compatible with 802.11. This was done so the many companies with existing 802.11 networks could more easily achieve migration. Moving from an 802.11 network to an 802.11b network was possible because of this compatibility requirement. © 2006 by Taylor & Francis Group, LLC



83

The data rate in 802.11b was called out as a maximum of 11 MB, a large difference from the existing 2-MB data rate available in the 802.11b standard. This greater bandwidth has helped many companies take their wireless systems from inventory management applications to desktop computing. This also sparked a new way to market wireless as a lowcost competitor to wired networks. With these incentives, many organizations upgraded their systems to support faster data rates. The 802.11b standard defines the way wireless networks operate at the physical layer and the logical or data-link layer. In the 802.11b physical layer portion, only a single spread spectrum technique was used. This is unlike in 802.11, in which there is a choice between Direct Sequence Spread Spectrum (DSSS) and Frequency Hopping Spread Spectrum (FHSS). DSSS is what 802.11b uses as a physical layer transport. This means that 802.11b uses a single preset 20-MHz channel operating in the 2.4 ISM frequency range when transmitting. DSSS was chosen over FHSS because it was determined that FHSS would be difficult to expand beyond the 2MB range. Now, looking at the MAC layer of 802.11b, one uses the Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) as a layer two transport. This portion of the 802.11b standard handles the movement of frames on a network segment. Frames control the flow of information and carry data. When creating the 802.11b standard, the IEEE had to define how frame management would take place and subsequently created CSMA/CA. The 802.11b standard is one of the most widely used wireless standards today. Most wireless systems are using this standard. Most all laptops ship today with built-in 802.11b or 802.11g wireless capability.

6.6.4 802.11c This was an older IEEE standard group working on MAC bridges. This working group created the 802.11c standard, which details how wireless bridges operate. Almost all the details of this standard only apply to designers and developers of the access point or bridge itself. Most installers or wireless network designers need not be familiar with the 802.11c standard other than to know it exists and that it details bridge operations.

6.6.5 802.11d The IEEE completed the 802.11d standard in 2001. It addresses the need for access points to have the ability to inform client cards of what regulator domain they are located at and what rules apply for that location. As seen in Chapter 4, many rules exist detailing the operation of wireless networks.



84


These rules only apply to the country in which the equipment is operating. In an effort to make interoperable products between these rules, a method for identifying what rules apply when became necessary. This is where the 802.11d standard comes into play. This standard gives business travelers the ability to use their wireless network cards in North America one day and in Europe the next day. The client card manufacturers also encouraged this standard. They wanted to provide a means to produce a single client card that would be able to carry over into other regulator domains. The 802.11d standard helps these manufacturers by not requiring them to create different client card versions for each domain in which they are going to sell them. This also helps normal business travelers because they do not have to carry around multiple client cards.

6.6.6 802.11e In the 802.11e standard, one sees the quality of service. This is critical in time-sensitive communications such as voice or video. Having a wireless network, QoS aware network is required to make quality calls on cordless VoIP phones. Most of the Wi-Fi phones on the market today use a proprietary method of QoS. Currently, the 802.11e standard is in draft. Once the draft becomes finalized, these proprietary mechanisms of QoS will adopt 802.11e.

6.6.7 802.11f This one standard should have been inserted directly into all 802.11 flavors. This has always been an amusing area in the wireless industry. When the 802.11 standard came about, interoperability was finally here. If a product was built to the 802.11 standard, it should work with other 802.11 products. This only held true when one meshed two access points made from the same manufacturer. If one took two products made by different manufacturers and tried to create one network, there would be no interoperability. The reasoning behind this was different roaming algorithms. Each manufacturer had determined a different method of roaming. It was not their fault that the 802.11 standard called for required support for roaming; unfortunately, they never defined how in the standard. To note here; most people who create a wireless network from two manufacturers think that their clients are roaming, when in fact they are going through the entire process of disconnecting and connecting to each manufacturer’s access point. The big difference here is time and state. In a roam, the time a client is not connected to the network is very minimal, so minimal that




85

it most likely will not interrupt any applications running. In the event that the client actually disconnects and reconnects, many applications will lose connectivity and produce various issues. The 802.11f standard provides a standard for roaming. This allows companies to create products that can seamlessly roam from one to another. This standard was championed by Cisco systems and was finalized in 2003. It specifies a protocol called the Inter Access Point Protocol (IAAP), which is a set of defined rules for how an access point should hand off clients and facilitate roaming. The IAAP is a multicast-based protocol that shares wireless connection information with all other access points on the same subnet. The ability of IAAP to move off a subnet is not possible due to a missing TTL field. Thus far, not many other companies have created 802.11f-compliant products, even with the standard fully ratified.

6.6.8 802.11g In 2003, the IEEE approved the 802.11g standard. When 802.11a came about, the IEEE recommended using a physical layer standard called OFDM. This allowed for a single channel that could be spilt it into multiple sub-channels, providing the capability to have greater data bandwidth on the same amount of frequency bandwidth. When 802.11g was conceived, the IEEE chose this OFDM technique for physical layer transport. This physical layer transport was applied to the 2.4 ISM, unlike 802.11a, which applied it to the UNII spectrum. This was done to be backwards compatible with the 802.11b standard. This means that the 802.11g wireless standard is capable of a bandwidth of 54 MB and can support 802.11b clients. Along with the added bandwidth of 802.11g comes a small loss in cell size compared to 802.11b. The 802.11g standard is very close to 802.11b from a coverage aspect; however, to get the two within ten feet, one needs to adjust the power settings of each radio. Why would one need to get these two standards so close? Most wireless devices today are still using the 802.11b standard. Although most laptops are moving away from the 802.11b standard and coming straight from the manufacturer with built-in 802.11g cards, they are not the most widely used wireless devices. Hand-held scanners, fork truck terminals, and other inventory management devices have wireless laptops, tablets, and PDA units outnumbered. This is because of their wide usage and deployment since the 802.11 days and even before that running on 900-MHz systems. Looking at B/G (i.e., 802.11b/802.11g) network coexistence, there are a number of limitations when conforming to the standard. All 802.11b wireless access points run in the same frequency as the 802.11g access points. This means the radios themselves are capable of running 802.11b © 2006 by Taylor & Francis Group, LLC


86


or 802.11g because they are both in the 2.4 spectrum. However, this does not mean an access point can switch between 802.11b and 802.11g without any hardware changes. The radio portion of the access point just propagates the signal to the antenna, regardless of how it is structured. A new or updated chipset is needed to change from DSSS to OFDM. This must be installed into any access points one wants to upgrade. Remember that all wireless access point vendors have different methods of upgrading, or, in some cases, replacement. One might need to purchase a simple expansion card or one might have to change out the entire access point. The 802.11g standard has a requirement to be backwards compatible. This means 802.11g must be backwards compatible with 802.11b. This backwards compatibility comes with a cost to network performance. When looking at B/G coexistence, one finds that when introducing an 802.11b client to an 802.11g infrastructure, it will default to an 802.11b wireless network. This means that if one allows 802.11b clients into an 802.11g network, it will lower the available bandwidth to 802.11b levels. This is because the access point has only one radio supplying two different physical layer techniques to clients. This was created to help companies move from 802.11b to 802.11g. Once all clients are 802.11g, the 802.11b data rates can be turned off and only 802.11g clients will be able to connect to the network. Until then, the 802.11g network most realistically always operates at 802.11b data router.

6.6.9 802.11h The 802.11h standard is looking at using 802.11a and developing the ability to self-tune, and moving away from congested channels. This will allow for easier setup and less administration. It plans to accomplish this by incorporating two techniques: (1) Dynamic Frequency Selection (DFS) and (2) Transmit Power Control (TPC). As one can see from the names, DFS changes the frequency or channel of an access point in the event that there is interference on that channel. The TCP portion of 802.11h controls the power output on each access point. This will allow each access point to know if it needs to increase power or lower power based on what it can hear from other access points in the area. These two techniques each handle changing the wireless channel or power in the event that interference occurs. This is being done today with products whose only purpose is to perform this type action. The IEEE would like to incorporate this functionally into a standard that each access point would utilize. The 802.11h standard is a move that the industry is making to create wireless networks without needing to design.




87

6.6.10 802.11i The 802.11i standard is a security standard that can apply to other 802.11 standards. It involves many changes, including the addition of advance encryption cipher AES and better key management functionality. After the many problems with the security built into the 802.11/a/b/g standards, a new security method needed to be developed. The IEEE created a group and finalized the 802.11i standard in 2004. For more information and details, refer to the 802.11i section in Chapter 12. For now, understand that the 802.11i is a security-related standard that runs on top of 802.11b/a/g, providing a more secure access method than the security defined inside each of these standards.

6.6.11 802.11j This standard is for use in Japan only. It defines the physical and MAC layer communications for systems running in the 4.9- to 5-GHz range. As of early 2005, the standard is still in draft format, with the working group still finalizing it. One interesting thing about this standard is the fact that North America has utilized this frequency exclusively for public safety networks. This may lead the 802.11j standard to become widely used in North America for police, fire, and other public safety needs. The United States is watching how Japan will use this standard and will evaluate it for public safety use in the United States.

6.6.12 802.11n Looking forward, the IEEE has started working on the 802.11n standard, which will be the replacement for 802.11g when it comes out. One interesting thing is the requirement that has been thrown around to support both 802.11g and 802.11a with the same device. Since the two standards are in two separate frequencies all together, in order to conform to the standard the access point will most likely have dual radios. Most of the Pre-N shipping equipment today is 802.11b/g backwards compatible and not 802.11a. This standard has been identified to operate at a minimum of 100 Mbps. To achieve faster data rates a new technique known as Multiple in Multiple out (MIMO) has been chosen for use. This works by having multiple signals sent and multiple signals received. In Chapter 8, we will talk about antenna diversity displaying and why this helps RF communications. MIMO goes further than antenna diversity by not only providing multiple antennas for receiving but multiple antennas for transmitting.



88


Wireless Data Rate Testing 20 18 16

5.5 Mbps

Speed

14

11 Mbps

12

24 Mbps

10

36 Mbps

8

48 Mbps

6

54 Mbps

4 2 0 802.11b

802.11a Standard

802.11g

Figure 6.5 Real wireless data rates.

6.6.13 Real-World Wireless Data Rates In looking at the many wireless IEEE standards, one sees that each standard outlines what their related data rates are. This number is not possible on normal networking protocols; they are roughly the top speed possible to send simple information through the device. The overhead of intelligent information that keeps track of the transmission takes overhead that the manufacturers and IEEE did not want to use in their top-speed testing. This means that none of the overhead of IP and TCP or UDP is part of this number. Most people will refer to this number as marketecture. Real-world testing reveals a major data rate difference from what is commonly relayed. Figure 6.5 outlines the marketed data rates and the real-world data rates. These data rates have been recorded in a computer laboratory under optimal conditions. This should make the real-world data rates as close to the real maximum data rates possible. It would be difficult to have a better connection in the field than in a controlled environment such as a laboratory. Another interesting fact is the amount of actual data rate gained between the 36-MB and 54-MB data rates; as one can see, this is very small. In some cases, the advantage of the 54-MB cell’s data rate outweighs the overhead required to support data rate roaming. What this means is that, in many cases, the 54-MB data rate is actually turned off to prevent excessive, unneeded roaming between the 48-MB and 54-MB data rates. © 2006 by Taylor & Francis Group, LLC



89

6.7 Personal Area Network (PAN) 802.15 The fight has been on for quite some time now as to how electronics makers can get away from cables. We have seen many cable types used to try to make the interconnection of devices as easy as possible. Take the USB, for example; this cable can be used for so many different things that an entire book could be devoted to it. For some time now, the push has been on for electronics makers to lose the wires. There is no reason why anyone would need to hook his or her radio up to a TV or a digital camera to a computer with wires. Using WPAN (wireless personal area network) technologies allows for the removal of cluttering wires, thus creating wireless interconnections to all support devices. The market has started to head in this wireless cabling direction for many devices. In doing this, many manufacturers are trying to develop a solution to get rid of the cables. As a direct result of this, there are currently a number of WPAN technologies. This section explores some of the most commonly known standards that make up these technologies.

6.7.1 Bluetooth 802.15.1 Bluetooth is a specification for wireless personal area networks (PANs) formalized by the Bluetooth SIG in 1999. It was originally developed by Ericsson, who was a member of SIG with IBM, Intel, Nokia, and Toshiba. The protocol operates in the license-free ISM band at 2.4 GHz, which is globally considered a free, open frequency band. It reaches speeds of 723.1 K a second and has a range of ten meters. It operates on Frequency Hopping and Time Division Duplex hopping 1600 times a second. Devices that are Bluetooth enabled will communicate with each other and find out if they need to be on the same network or not. Once two or more devices realize that they need to be on the same network, a piconet is formed. They know this by a predefined address setting either created by the user or input into the device by the manufacturer. This piconet is used to synchronize each participating device so that they use the same hopping sequence. Some of Bluetooth’s more predominant uses include wireless headsets for mobile phones. These can save time and lives while in use in a car. Another common use of Bluetooth is for wireless keyboards and mice. Recently, new uses for Bluetooth have emerged. One such use is to replace wires in home theater systems. Some way-out uses for Bluetooth that may become commonplace in the near future include some initiatives that Motorola is currently working on. The first one is a partnership with snowboard and clothing manufacturer Burton. Motorola wants to include Bluetooth-enabled speakers and microphones in Burton clothing and © 2006 by Taylor & Francis Group, LLC


90


helmets. This would allow snowboarders to listen to an MP3 player or use a cellular phone located deep in their pockets with their helmet. Placing a speaker in the helmet and a small number of push-buttons on the sleeves of clothing would allow these Bluetooth-enabled pieces of clothing to place calls and listen to music from any Bluetooth-enabled device. Another partnership that Motorola has started is with sunglasses manufacturer Oakley. The sunglasses will allow the user to listen to MP3 and take cellular phone calls from any Bluetooth-enabled device.

6.7.2 Infrared (IR) Infrared (IR) is not considered a WPAN technology and is not part of the 802.15 family. It is listed here because it has a very small distance limitation, which qualifies it as a personal area network (PAN). Interestingly enough, IR was originally defined in the 802.11 standard. The standard IR has allowed us to connect many devices together quickly and easily for small file transfers. IR is widely deployed on laptops, cell phones, PDAs, and many other hand-held mobile devices. Many people have looked at the security risk of IR; although it has very limited range with almost virtually no security, the risk is low. The risk is so low because of the range factor; one needs to be within one or two meters of another device. This allows anyone to easily see someone who would be trying to attack his or her IR device.

6.7.3 Ultrawide Band 802.15.3 The ultrawide band (UWB) is very similar to how it sounds. A short-range communication technology transmits its signal across a wide frequency range. This wide range is also where other signals would be located. UWB is able to prevent itself from affecting or being affected by these other signals by allowing such a wide band. As discussed in Chapter 4, narrowband communications are easily detected and avoided when transmitting a signal at low power over a wide frequency. This is the goal of UWB communications. Currently, there are ongoing talks concerning uses for UWB. Companies have started to develop products used for data transmission, location, and even warfare. The military has been interested in UWB because of its low power and resilience to jamming. Currently, the UWB standard is being headed up by the IEEE. They have created the IEEE 802.15 WPAN High Rate Alternative PHY Task Group 3a (TG3a) to standardize UWB. This standards body currently has two proposed standards for UWB, and is in the process of removing one of them. Once a choice has been made, that version will become the true



Wireless Local and Personal Area Networks 91

UWB standard. Some of the details of each of these proposed standards are listed below. For a quick overview of what to expect from UWB, one can count on very high data rates, upwards of 500 Mbps standard. One standard has UWB capped at 1320 Mbps and the other has it capped at 480 Mbps. Both will use the unlicensed 3.1- to 10.6-GHz UWB. How the modulation is handled is where each of the two proposed standards have the most differentiation. One calls for using direct sequence spreading of binary phase shift keying (BPSK) and quaternary bi-orthogonal keying (4BOK) UWB pulses. The other calls for using orthogonal frequency division multiplexing (OFDM) with 122 sub-carriers modulated using quadrature phase shift keying (QPSK). Either way this standard goes, it will be a long-awaited replacement for electronic device cabling.

6.7.4 ZIGBEE 802.15.4 Zigbee was created from the 802.15.4 standard. This standard was published in May 2003 to create a low data rate solution that could have long battery life. The battery life is dependent on the device it is servicing. The 802.15.4 standard calls for multi-month to multi-year battery life. Another key function of the standard is its ability to operate with very little complexity. The 802.15.4 standard, like most IEEE standards, only addressed the physical and MAC layer portions. The Zigbee Alliance came in and addressed a standard for the network layer through the application layer. This included security management, topology management, routing, MAC management, discovery protocol, and in open API for vendors to create their own unique features and functions. Having this API only available at the application layer of the OSI model allows the Zigbee standard to have interoperability between multiple devices and multiple vendors. The Zigbee standard can operate in three frequencies: one worldwide, one outlined in Europe, and the last one available in the United States. The first worldwide frequency is 2.4 GHz. This is located in the ISM band, along with Bluetooth, 802.11, 802.11b, and 802.11g. This makes this frequency very vulnerable to interference from these other technologies or cordless phones and microwaves, which also run on this frequency. Some of the positive features of this frequency are worldwide adoption, high data rates, and the highest available channels. The Zigbee standard running in the 2.4-GHz ISM band yields 250 kbps with 16 available channels. This frequency uses a modulation technique that is different from the other frequency choices. It uses 0-QPSK. The next frequency is the one based in Europe; it is 868 MHz. This frequency has a much lower speed than if it was in the 2.4-GHz band. The current data rate on the © 2006 by Taylor & Francis Group, LLC


92


868-MHz Zigbee standard is 20 kbps, with only one static channel available. It runs using BPSK modulation. The last open frequency that can be used with Zigbee is the 915-MHz ISM band. This also uses the BPSK modulation, is used in the Untied States, and has a data rate of 40 kbps with 10 available channels.

6.8 Chapter 6 Review Questions 1. The 802.11 standard provides what maximum data rate? a. 11 Mbps b. 2 Mbps c. 5.5 Mbps d 54 Mbps 2. With Ad Hoc mode, an AP is required. a. True b. False 3. Most wireless networks operate in what mode? a. Ad hoc b. Infrastructure c. Repeater d. Bridge 4. The 802.11b standard provides what maximum data rate? a. 11 Mbps b. 54 Mbps c. 2 Mbps d. 1 Mbps 5. What part listed below radiates a wireless signal? a. Access point b. Antenna c. Network card d. Wireless laptop 6. Which of the following terms do 802.11/b/g/a have in common? a. CSMA\CD b. CSMA\CA c. DSSS d. OFDM




93

7. In the United States, what is the maximum EIRP limit on 802.11a UNII 1 system? a. 200 mW b. 100 mW c. 250 mW d. 300 mW 8. The 802.11a standard provides what maximum data rate? a. 11 Mbps b. 2 Mbps c. 48 mbps d. 54 mbps 9. What UNII band is for outdoor use? a. UNII 1 b. UNII 2 c. UNII 3 d. UNII 4 10. The 802.11g standard provides what maximum data rate? a. 11 Mbps b. 48 Mbps c. 24 Mbps d. 54 Mbps 11. Which of the following PAN technologies does 802.11 address in its standard documentation? a. Bluetooth b. IR c. Zigbee d. UWB 12. What happens on an 802.11g network when an 802.11b client is injected? a. Nothing, the client will not operate. b. The access point will service both clients without any problems. c. The access point and all other access points that can hear the client will down-shift to 802.11b. d. The access point that the client is connected to will downgrade to 802.11b to support that client.



94


13. What standard would one need to install Wi-Fi phones in a network? Select the best answer. a. Wi-Fi b. VoIP c. 802.11e d. 802.11g 14. What is the newly formed draft standard that will replace 802.11g? a. 802.11b b. 802.16 c. 802.11n d. 802.11z 15. Which one of these terms is not part of the wireless local area network service set? a. BSS b. IBSS c. ESS d. IESS 16. Roaming is defined in which of the following standards? a. 802.11 b. 802.11g c. 802.11b d. None of the above e. All of the above



Chapter 7

Wide Area Wireless Technologies There are many types of RF networks throughout the world. Just sitting there reading this, there are hundreds of wireless signals bouncing off your head — radio, TV, cell phones, GPS, and lots of governmentcontrolled signals. Chapter 1 explored some of the inherent problems facing all wireless signals; this chapter looks at how companies have combated these problems in the past as well as the present. One of the biggest issues with wireless falls under the frequency spectrum. Creating how the system will work is not difficult; however, getting the OK from national governments to use certain frequency spectrums is difficult. If a non-government-licensed spectrum, such as one that is free to use, is adapted by a company, providing reliability becomes a challenge. Security is of major concern for most of these signals. TV and radio are not very concerned about security because they do not hide their existence but rather publicize it with commercials and advertising. A lot of the signals discussed in this chapter, such as cellular phone calls, are very sensitive and require some means of security. Looking at how security was and is deployed for these different types of RF technologies gives us a better idea of how we deploy security within our WLAN (wireless local area network).

7.1 Cell Phone Technologies Cell phone technology has seen three generations of change since its major commercial debut in 1980. Looking at each of the generations of 95 © 2006 by Taylor & Francis Group, LLC


96


technology up close, one notices the advancement of security throughout the cycle. With first-generation (G1) cell phones, we had unencrypted analog voice transmitted in plain FM. With second-generation (G2) technology, we had encrypted communication and better quality. Third-generation (G3) has us using phones to make purchases that are deducted directly from a credit card or even used to carry a line of credit with a cellular provider. Third-generation also prompted the “be connected everywhere” communication phrase. To support this requires the availability of data access technologies on mobile devices. Third-generation cellular created this data access, although it was low data rate access. This low rate access prevents advanced features. People have already started to look at fourth-generation wireless. Using a high-speed network capable of providing the bandwidth needed to host multimedia-intense applications is just a pipe dream today. The only real solution conceivable would be a fourth-generation network made up of multiple networks using multiple technologies leveraging the fastest available delivery method. This means a wireless end device would be able to connect to multiple technologies, making decisions on what network to use based on highest bandwidth available from any carrier’s networks. One might picture a fourth-generation (G4) network as follows. As you leave your house in downtown Manhattan, your wireless device jumps from your home network to a GSM or other cellular network. Then, once in range, you might roam into one of your carrier’s 802.11a/b/g/n participating hotspots or even across a longer-range 802.16e network. After you leave the congested city, your available choices as well as your bandwidth would be greatly reduced because the presence of smaller cell size, higher bandwidth networks decreases. Looking at Figure 7.1, one can see an example of how a G4 network operates.

7.1.1 Analog Analog phones were the first type of cellular phones released. These phones were insecure, with voice traveling over the airwaves without any encryption. These networks were often small and suffered major interoperability problems. After the industry found this major interoperability problem, it decided to create the Advanced Mobile Phone Systems standard (AMPS). When companies started to make their networks conform to this standard, they had few interoperability problems. This standard also allowed devices from various manufacturers to work on each other’s wireless networks, thus opening up the market for more competition. This competition led to more device choices and happier customers. Looking at the security of analog cellular networks, one notes that all conversations taking place were unencrypted. This meant that with a © 2006 by Taylor & Francis Group, LLC


Wide Area Wireless Technologies

97

GSM CDMA Cell Phone Tower

Large 802.16 wireless networks

Large 802.16 wireless networks

Small 802.11b/g/a hotspots

GSM CDMA cell phone tower Small open wireless project participants

Figure 7.1 Fourth-generation network diagram.

simple police scanner, one could pick up and listen to anyone’s calls. This became worse when criminals learned how to copy or clone these phones by extracting their identification information and copying it to a different phone.

7.1.2 TDMA In the late 1980s, the wireless industry began to look at converting its analog networks to digital networks. This was to improve capacity, security, and quality. In 1989, Time Division Multiple Access (TDMA) was chosen over Motorola’s Frequency Division Multiple Access (FDMA) by the Cellular Telecommunications Industry Association. This decision helped TDMA become widely used and deployed. © 2006 by Taylor & Francis Group, LLC


98


Frequency 1.0 GHz

Phone 1 P2

P1 P3 P2

P1

P3

P2 P1

P3

P2

Phone 2

Phone 3

850 MHz

Figure 7.2 Time division multiple access (TDMA).

TDMA (Figure 7.2) is digital cellular transmission technology. It allows for a reduced amount of interference by allocating unique time slots to each user within each channel being used. This allows a number of people to access the network and make a call using a single RF channel. The TDMA digital transmission divides a single channel into multiple time slots. It works by taking audio signals and digitizing them into short, millisecond-long packets and then evenly spacing these inside a single channel. This takes place for a short time and then the entire channel is moved in frequency and the whole process starts over. TDMA offers a number of other advantages over analog cellular technologies. It can carry data rates of 64 kbps to 120 kbps. This allowed many other uses of this technology, such as fax, Internet access, short message service, and text messaging, as well as other multimedia application uses. TDMA also provides cell phones with extended battery life and talk time. This is due to the amount of transmitting actually taking place. A TDMA phone is only transmitting one third to one tenth of the time during any given conversation. Of course, there are some disadvantages to TDMA technology. When roaming from one wireless cell to another, a call could be dropped. This is because there are no means of time allocation on the new cell calls. Once you roamed onto a new cell and the cell did not have an open time slot to insert your call, it was dropped. If all the time slots in the cell were already occupied when a user tried to redial the dropped call, they may not have received a dial tone.




99

Frequency 1.0 GHz

Phone 1

010110 010110 010110 010110 010110 010110 Phone 2

Phone 3

850 MHz Codes known by cell phones

Codes known by cell tower

Phone 1 code = 4th and 2nd bit Phone 2 code = 5th and 1st bit Phone 3 code = 6th and 3rd bit

Phone 1 code = 4th and 2nd bit Phone 2 code = 5th and 1st bit Phone 3 code = 6th and 3rd bit

Figure 7.3 Code division multiple access (CDMA).

7.1.3 CDMA Code Division Multiple Access (CDMA; Figure 7.3) was originally developed for U.S. military use in the early 1960s with the help of a company called Qualcomm. The decision to develop this technology came from a need to have communication channels more resistant to interference caused by enemy jamming devices. Also required was the need to add some form of encryption to prevent eavesdropping. Because Qualcomm had led the development of the CDMA technology, they were the owners of a large amount of intellectual property, patents, and trade secrets pertaining to CDMA. In the early 1990s, the Telecommunications Industry Association adopted CDMA as a digital access technique for cellular phones, making a once military-only technology available to the public. This technology works by assigning a digital code to every bit of speech and then sending the encoded speech over the airwaves. All calls that are placed in the area of coverage of a single cell phone tower are placed within the same radio channel. This means that at any given time, there are numerous calls traveling the airwaves acting like a single signal. Each call is broken out from the transmission with sequential code. This code is only known to the receiving phone and the cellular tower. This code allows one to pick certain bits from the airwaves and reassemble



100


the call, thus allowing secure communications. Defeating this code is extremely difficult because there are over 4.4 trillion codes. This large number of codes prevents eavesdropping and protects confidentiality.

7.1.3.1 CDMA2000 The IS-2000 standard (CDMA2000 1X) was completed and published by the Telecommunications Industry Association in 2001. The CDMA2000 1X offers twice the voice capacity of the original CDMA or what is now known as CDMAOne with the advent of this new technology. It also provides average data rates of 144 kbps. With this increased speed and data access, cellular phones can utilize their cellular connection to provide services such as Internet access, weather, news, stocks, and e-mail. This has led more and more cellular customers to converge PDA devices and cell phones into a single device. Cellular carriers have rapidly adopted CDMA2000 due to its backwards compatibility with the original CDMA and the low cost of upgrading their networks to support it.

7.1.3.2 CDMA 1xEV-DO and CDMA 1xEV-DV Once CDMA2000 was available, customers quickly became interested in receiving more bandwidth. Because of this, CDMA was once again modified to increase its data speeds. This became known as CDMA2000 1xEV, which stands for 1x Evolution. Currently, CDMA2000 1xEV has two phases: (1) 1xEV-DO, which provides data only, and (2) 1xEV-DV, which provides data and voice. In 2002, 1xEV-DO systems were installed, and although they do not provide voice capabilities they are CDMA2000 1x backwards compatible. The current speeds of 1xEV-DO are in excess of 2 Mbps if the network is clear and voice calls are handled with another carrier. Because of backwards compatibility, there have been talks about having data and voice run on two different networks. This would allow allocation of maximum bandwidth to the data connection. As for 1xEV-DV, this enhancement will fix the bandwidth issue of running voice and data from the same carrier. To support these increasingly bandwidth-intensive technologies, the cellular carriers have started to converge their networks into IP (Internet Protocol).

7.1.4 GSM In 1982, the Conference of European Posts and Telecommunications (CEPT) created a GSM group (Groupe Spécial Mobile). Its job was to define a standard mobile network for use across Europe. In 1982, GSM did not © 2006 by Taylor & Francis Group, LLC



101

stand for Global System for Mobile Communications because it was only defined as a standard mobile network used across Europe and not the world. In 1989, the European Telecommunications Standards Institute (ETSI) took control of the GSM group and in 1990 finished the first GSM standard. One year later, GSM networks were starting to rise all over Europe. The GSM network was one of the first digital or generation-two networks at the time. Today, GSM is the most popular cellular technology in the world. GSM itself uses FDMA and TDMA as part of its access scheme. It is quite interesting how it works; GSM takes the available bandwidth and breaks it up into 124 sub-carrier frequencies using FDMA. There are 125 available carriers, although the first one is just to protect the GSM signal from interference. The remaining 124 sub-carriers are each divided into eight time slots using TDMA. This allows four devices to use each channel. This is because each device uses two channels — one to transmit and one to receive. With this, many people have the impression that the device is simultaneously sending and receiving. Because TDMA was used to divide up the channel, there are eight time slots, which means that a device using two of the eight channels is never using them at the same time. The setup of the GSM network is very involved; for example, the current GSM standard is more than 8000 pages long. Because of the depth of the GSM system setup, only a high-level discussion of how it works takes place here. In a GSM system, many parts are required. From a highlevel view, only a select few are explored herein. The first item is the phone; most often the phone is referred to as the mobile station and designated as MS. This device holds a SIM card, which contains the phone book and all the carrier subscription information for the phone. The phone also holds the MSISDN, which is a unique identifier that informs cellular networks who the original carrier is. This is used for finding the authentication and accounting information so the carrier can send out the cellular bill. The next device is the cellular tower. Once the phone connects to the network, the cellular tower, called a base station, receives the call. The base station then forwards the call to the base station controller. This device performs all the management of the call and relays this to the tower. This is similar to the access point and its antennas; the tower itself is only the antenna and the brains lie in the base station controller. Well before any calls take place, a number of control messages and exchanges take place. When a phone is turned on, it will try to find its location information. This can be requested by the phone or queried by the cellular network. Either way, the result is what is called a Mobile Station Roaming Number (MSRN). This number is based on location; this means that a select few of these numbers always remain in the same general physical location.



102


Visitor Location Register (VLR) Mobile Station (MS) Home Location Register (HLR)

Base Station

Gateway Mobile Gateway Mobile Switching Center (GMSC) Switching Center (GMSC)

Figure 7.4 Global system for mobile communications.

Now, how did the phone get its MSRN? It requested it by contacting the local Gateway Mobile Switching Center (GSMC). This is a device that handles all call processing. This is very different from the functions of the base station controller, which only handles the connection management of the mobile station to the base station or cell phone to tower conversation. Once the base station controller sees the phone’s request for its MSRN, it forwards the request to its local GSMC or gateway. The GSMC will use the MSIDN to find the carrier’s gateway or GSMC. Once the carrier’s GSMC is found, the other gateway will send it the Visitor Location Register (VLR) or one’s current location and receive the Home Location Register (HLR). Now both gateways know where the device came from and where it is now. Once this process is complete, a call can take place. Figure 7.4 shows how each of these pieces interacts with each other. The operations of a call are very similar to the control and location functions. The phone will send an “I want to call XXX-XXXX” to the base station. The base station will receive this and pass it along to its controller, which then in turn forwards it to the local gateway. The local gateway looks at the number and finds the calling number’s gateway. Once the gateway is found, a request is made for the HLR and VLR so that the MSRN can be found. What this means is that the gateway that is local to the phone’s current position tracks down the device, keeping track of the other mobile phone. Once the other gateway responds, it tells the current gateway where that number can be found. Once this is done, the gateway handling the call locates the other number and sends the call. As stated above, there are many other pieces to this, but this definition should help the majority of people understand how GSM works.

7.1.4.1 GPRS The General Packet Radio Service (GPRS) is a high-speed mobile information service. It allows packet-based communications to take place on © 2006 by Taylor & Francis Group, LLC



103

a GSM mobile network. This is very different from the circuit switched network that most cellular and telephone networks utilize. GPRS uses channels of GSM, or to be more specific, it uses time slots in TDMA. This allows for multiple mobile stations to use GPRS in a given area without filling up the voice channels. In these networks, voice always has priory over data. This means that in a highly concentrated cell, GPRS has low data rates. The data rate of GPRS is 170 kbps, although in real-world cases it is closer to 25 to 75 kbps. GPRS has a feature called classes. This allows the device to use different numbers of available channels for uploading compared to downloading. This would be relevant when the application using GPRS is one that is primarily receiving data compared to one that has equal sending and receiving. In either case, one would adjust the class feature to meet specific needs.

7.1.4.2 GSM Security System Overview The security system of GSM is even more complex than the network system. This section attempts to outline it as best as possible without going so in-depth that the technical information becomes overwhelming. The GSM security model is based on a shared secret that is located on the mobile phone inside the SIM chip and the subscriber’s HLR. This shared secret is a 128-bit key that is used to generate a 32-bit response. It is identified as Ki. When a mobile station first comes online, it connects to the local gateway based on its position. This gateway then tracks down the phone home gateway and receives a random challenge and signed response from its home HLR. This also comes with two other pairs of challenge and signed responses that are used later. After receiving both pieces of information, the local gateway will only send the challenge to the mobile station. When the mobile station sees this challenge, it will use its shared secret to create a signed response and send it to the local gateway. To create this response, it will use the A3 cryptographic algorithm with the shared secret and the challenge. Once the gateway sees this, it will compare it to the already sent signed response from the subscriber carrier. Now we have taken care of authentication; this has only gotten the phone onto the network and able to place calls. The next section explores how the calls are encrypted. Now let us return to the second pair of challenge and signed response messages. These messages were created with an algorithm other than A3. They use an algorithm called A8 to create a key that is used with another algorithm called A5 to encrypt the data. This works much the same way as authentication works. The mobile station uses the challenge and their shared secret programmed into the mobile devices to create a key. This © 2006 by Taylor & Francis Group, LLC


104


process uses another algorithm, called A8, to create a session key called Kc. The Kc key is used with a frame number to create a unique key stream for every frame. Another algorithm used is called COMP 128; it is used for both A3 and A8 algorithms in most GSM networks. The COMP 128 algorithm generates both the signed response and the session key in one run. The key length of COMP 128 is 54 bits instead of 64 bits, which is the length of the A5 algorithm key. Ten zeros are appended to the key as padding when generated by the COMP 128 algorithm. This means that the keyspace used to protect the key is not 64 bits, but rather 54 bits. Over the years since GSM came out, some attacks have been released that defeat this security method. A key note that needs addressing here is that the A3, A5, and A8 algorithms are not publicly available. It has always been correct procedure for any algorithm to have open source available to prove a good level of security by subjecting it to academic and professional scrutiny. This ensures that an algorithm is actually safe. The process of having an algorithm open has been default practice for all commercial, financial, and governmental usage for more than 20 years. The GSM algorithm for A3 and A8 uses the COMP 128 algorithm as well. This algorithm was leaked to the public; and once it was securitized, it was quickly noted that a problem exists. In 1998, members of the Smart Card Developers Association demonstrated that they could crack the A5 authentication method of GSM in a matter of hours on a single PC. This experimentation led to more in-depth analysis and later two men (Alex Biryukov and Adi Shamir) said they were able to perform an attack on A5 with a single computer in less than two minutes. Because of these threats, it is widely known that GSM systems have a poor security method.

7.2 GPS The Global Positioning System (GPS) is a U.S. military space system operated by the U.S. Air Force. The system is used for position location, navigation, and precision timing. It accomplishes this using three segments: (1) satellites, (2) ground control centers, and (3) receivers. The most common consumer application is giving real-time directions to drivers; there are thousands of popular government applications, along with some special public uses such as tracking. Satellites are the first and most widely known segment of GPS. These 24 orbiting satellites are stationed 11,000 miles above the Earth. These satellites send one-way, time-tagged transmissions that radiate down over the entire Earth. This timing is produced by four atomic clocks located




105

on each satellite. The Department of Defense officially named the satellites NAVSTAR, which stands for Navigation Satellite Timing and Ranging. These satellites were launched in phases, with the first phase of four satellites going up in 1978 and the final phase completing full global coverage 17 years later in 1995. The second segment of GPS is the control segment, consisting of five tracking stations located across the world. The master control center is located at Schriever Air Force Base, in Colorado. This is home to the Air Force’s 50th Space Wing, the unit that provides command and control for defense warnings, along with navigational and communicational satellite tracking. The other four control centers are unmanned and controlled from the master control center. These four control centers are located in Hawaii and Kwajalein in the Pacific Ocean, Diego Garcia in the Indian Ocean, and Ascension in the Atlantic. The location portion runs across the GPS, but all tracking services utilize some other technology to transmit the GPS coordinates. It is important to note that GPS by itself is not a tracking system. GPS receivers use the passive, one-way transmissions of signals from the orbiting satellites to determine a position fix. Nothing in the GPS infrastructure enables system operators to know who is using the signal or where they are. Conversely, a GPS unit cannot use the system to transmit its own location. For GPS to become part of a tracking system, it must be coupled with a communications device. Network-based positioning systems that use a telecommunications infrastructure can determine a user’s location. This location information is either available on the device or needs another technology like cellular to transmit. A GPS-equipped digital phone can also be used to track its location and movement. These phones provide users with greater control over their own privacy because such units typically allow users to block the transmission of the phone’s location information. Looking at GPS security, one can start with its public usage debut. During President Reagan’s term, GPS was made freely available worldwide for commercial use. When the system was released, it was set up to beam down two signals: (1) the Precise Positioning Service (PPS) or Precision code (PC), and (2) the Standard Positioning Service (SPS) or the Coarse Acquisition code (CA). The dual signals arose from a national security concern of allowing near-perfect positioning to enemies. This PC code is the one the military encrypts with confidential cryptographic equipment, keys, and specialized receivers. The PC code also increases its accuracy by using a second carrier wave that allows receivers to measure the small delay caused by the signals having to move through the atmosphere. The other code (the CA code) was the one released to the public. It was less accurate and easier to jam. It is easier to acquire the CA code than the



106


PC from the Earth; it is so easy that the military first homes in on the CA code. Once it has found and tracked the CA code, it authenticates using their cryptographic key, allowing them to transfer to the PC code. This CA code is defused by a technique called selective availability (SA). This technique allows the U.S. Government to modify the GPS signal accuracy in any place on Earth at any time. This selective availability could be used by the U.S. Government to change the course of an enemy military relying on GPS for navigation. Although the Department of Defense tried to limit the accuracy of commercially available GPS, the GPS receiver companies soon found legal ways around selective availability. This technique, which had major involvement from the U.S. Coast Guard, is called Differential GPS. The technique worked by using the exact position of the base stations and comparing it with the base station’s location through the GPS CA code. Once the level of error was determined, it could be applied to the location of the receiver, making it almost dead accurate. Well after a number of techniques similar to the one described above were created to get around the selective availability, a major event took place. This event changed all of these concepts and techniques. On May 1, 2000, President Clinton signed an order ending selective availability and making civilian GPS readers a lot more accurate. This led to the commercial end of techniques to get around selective availability, and also created the massive proliferation of commercial GPS devices. There have been many U.S. Government documents, reports, and appointed teams tasked with researching vulnerabilities in the GPS. Most of this work leads to the notation that the main risks in GPS are signal jamming and spoofing. Looking at what kind of jammer one would need depends on the amount of space in which one wants to prevent the GPS signals. To have any real effect on the system, one needs a jammer capable of jamming up to 10,000 watts. This would mean one would need a large truck or plane. Luckily, for the military, the bigger the jammer, the easier it is to track down and destroy. The real threat to GPS comes from a large number of small, man-portable, 500- to 1000-watt jammers. With these jammers running simultaneously and spread out in a large area, an enemy could easily create a large GPS jamming capability. For this reason, the use of small wideband jammers is of great concern to the U.S. military.

7.3 802.16 Air Interface Standard The IEEE 802.16 Air Interface Standard is for fixed broadband wireless access systems employing a point-to-multipoint architecture. The initial




107

version was developed with the goal of meeting the requirements for deployment of systems operating between 10 and 66 GHz. This has made the standard extremely large in scope, so much so that an amendment was made (called 802.16a) to finish the standard for systems operating between 2 and 11 GHz. One of the main reasons that the 802.16 standard is not taking over the DSL and cable market share is the high cost of customer premise equipment (CPE). Also in 802.16, no mobility is supported. Recently, the battle, primarily headed up by Intel, created the requirement to support mobility in the 802.16 standard. This did not make it into the 802.16 standard, although it did create the 802.16e draft standard. This could make some major changes to how people use WLAN technologies. With 802.16e, which allows laptops or other wireless devices to connect to networks that can service many miles, we may start to see pure cellular technologies disappear. The IEEE process stops short of providing conformance statements and test specifications. To ensure interoperability between vendors competing in the same market, the WiMAX technical workgroup was created by the leaders in IEEE 802.16 technology. WiMAX is the tradename for the IEEE 802.16 standard for high-speed metropolitan area wireless networks that are interoperable. These products are expected to come to market in 2005/2006.

7.4 802.20 Standard The 802.20 standard is also being created for long-range wireless access. This standard is designed to address the need to receive signals traveling at a high rate of speed, such as 150 miles per hour. This would be useful for high-speed trains. One of the ideas behind 802.20 is to have a large infrastructure where a subscriber could have an always-on connection. This would take Wi-Fi and cellular to new levels where high-speed data devices would be located in everyone’s pocket and be as common as a cell phone. This standard is in the draft stage and is still very new; there is only limited information as to how it will work and at what frequency. Right now, it is looking at using an open bandwidth around the 3.5-GHz range. Frequency range is a big issue for long-range wireless solutions; using open-band, freely available spectrums means large amounts of unavoidable interference and government-regulated frequencies require a license and a hefty fee.



108


7.5 Chapter 7 Review Questions 1. What type of cellular communications is considered G1? a. Digital b. GSM c. AMPS d. CDMA 2. What type of cellular communications is considered a global standard? a. TDMA b. GSM c. AMPS d. CDMA 3. What is the data rate of the CDMA2000? a. 21 Mbps b. 144 kbps c. 550 kbps d. 2 Mbps 4. The 802.16 standard has built-in mobility. a. True b. False 5. GPS has how many satellites orbiting the earth? a. 24 b. 26 c. 28 d. 30 6. What cellular standard was held back for many years as a militaryonly technology? a. TDMA b. GSM c. AMPS d. CDMA 7. Which of the following terms is the one that finds other cellular users? a. Gateway mobile switching center (GSMC) b. Home location register (HLR) c. Visitor location register (VLR) d. Mobile station roaming number (MSRN) © 2006 by Taylor & Francis Group, LLC



109

8. What is the documented as well as the real data rate of GPRS? (Select two) a. 170 kbps b. 210 kbps c. 10 kbps d. 55 kbps 9. What is the data rate of the 1xEV-DO? a. 21 Mbps b. 100 kbps c. 550 kbps d. 2 Mbps 10. Which cellular standard uses slices of time to allow for multiple connections? a. TDMA b. GSM c. AMPS d. CDMA 11. The A3, A8, and A5 algorithms are part of what standard? a. TDMA b. GSM c. AMPS d. CDMA 12. All the research performed on the security of GPS has led the U.S. Government to say that the biggest risk to the GPS system is ________. a. Security b. Jamming c. Spoofing d. Attacking the satellites 13. The 802.16e standard addresses mobility. a. True b. False 14. What body performs interoperability testing for 802.16? a. WECA b. Wi-Fi Alliance c. WiMAX d. Wi-FiMAX © 2006 by Taylor & Francis Group, LLC


110


15. What is the major difference between 802.16 and 802.20? a. Range b. Speed c. Ability to receive a signal at high rates of velocity d. Ability to securely connect to a wireless network 16. What is the data rate of TDMA? a. 2 Mbps b. 100 kbps c. 64 kbps d. 500 kbps



Chapter 8

Wireless Antenna Theory All wireless devices need an antenna to send or receive signals. The antennas are what the devices use to find or send wireless signals. Without an antenna, a wireless device would have virtually no means of achieving any type of wireless communication. Placing an incorrect antenna type could dramatically reduce the usefulness of a wireless network. This chapter familiarizes the reader with antennas by looking at how they work. Given a basic understanding of the antenna properties, this chapter then explores the different types of antennas as well as the types of connectors used. It also looks at homemade antennas — where they come from, what they are made of, and what laws exist pertaining to them.

8.1 RF Antenna Overview Understanding what kinds of antennas are available as well as how they perform is a critical part of understanding wireless networking. To place antennas properly, an understanding of how they work is required. This will ensure that incorrect antenna placements do not degrade the wireless infrastructure. Given a full understanding of how antennas work and what types of antennas radiate signals in what ways, this chapter then explores which antennas are best for particular situations.



112


Isotropic radiator Antenna radiation

Horizontal

Vertical

Figure 8.1 Antenna polarization.

8.1.1 Polarization Radio waves are created by two fields: (1) electric and (2) magnetic (Figure 8.1). These two fields are perpendicular to each other, allowing energy to flow from one to the other. This energy flow is what is known as “oscillation.” The position and direction of the electric field with reference to the Earth’s surface determines the wave’s polarization. The metal portion of the antenna that is actually radiating the signal is always parallel to the electric field. (This means that an antenna that admits an electrical field horizontally is polarized the same.) Understanding the correct polarization of an antenna can help avoid incorrectly mounting it and causing a coverage issue. Horizontal polarization: the electric field is parallel to the ground. Vertical polarization: the electric field is perpendicular to the ground.

8.1.2 Gain Antenna gain is measured in dBi, which gets its value from a comparison against an isotropic radiator. An isotropic radiator is a sphere that radiates power equally in all directions, both vertically and horizontally. When we look at antenna radiation pattern against that of an isotropic radiator we see a pattern more closely related to a doughnut than a perfect sphere. When a greater gain is needed, we push two of the vertical ends together, creating a larger horizontal gain. This is similar to pushing two sides of a balloon together. When this is done, the balloon protrudes out the



Wireless Antenna Theory

113

opposite direction that it was pushed. Both dBi and dBd are used to measure the gain on antennas. When confronted with a measurement of dBd, one can simply add 2.14 to the dBd number to produce its dBi counterpart. This is essential for an RF engineer to understand. An engineer would need to know this because, as discussed in the next section, dBi is used to find the power limitation created by the FCC called EIRP (equivalent isotropic radiated power). Most of the time, antenna manufacturers list their antennas in dBd because it is a larger number and often makes their antennas look more powerful to uneducated customers.

8.1.2.1 Equivalent Isotropic Radiated Power (EIRP) A term used to measure the transmit power is called the equivalent isotropic radiated power (EIRP). EIRP represents the total effective transmit power of the radio. This includes any gain that the antenna provides, as well as any losses from the antenna cable. EIRP is the total wireless signal power output on a radio device. The EIRP value is regulated by the FCC (Federal Communications Commission), meaning that only allowed values can be used and sold. These rules and values are listed in FCC Part 15.247. All manufacturers of 802.11b/a/g products must comply with this to be able to sell their products. An RF designer must be able to measure EIRP. To do so, the following formula must be applied: Transmitter power + Antenna gain − Antenna loss = Equivalent Isotropic Radiated dBm dBi dB Power (EIRP)

EIRP is calculated by adding the transmitter power to the antenna gain and then subtracting that from any cable loss. It is expressed as transmitter power (in dBm) plus antenna gain (in dBi) minus antenna loss ( in dB). Therefore, to go through an example: a radio with a transmitting power of 20 dBm using an omni antenna, with a 5.2-dBi antenna gain connected to a low loss cable. The low loss cable is 10 feet long and injects a total loss of 5.5 dBi. To solve this: 20 + 5.2 − 5.5 = 19.7.

8.1.3 Beamwidth Beamwidth is the width of an antenna’s gain. When one narrows or focuses an antenna’s beam, the gain will increase. There are two factors associated with beamwidth: (1) vertical and (2) horizontal. The vertical beamwidth is measured in degrees and is perpendicular to the Earth’s surface. The horizontal beamwidth is measured in degrees parallel to the Earth’s surface.



114


Path loss More signal

Less signal

Figure 8.2 Path loss.

Using beamwidth when designing a wireless network is very important and one must understand what applications require a greater horizontal rather than a vertical beamwidth, or vice versa. When trying to cover a long narrow hallway, using an antenna with a greater vertical beamwidth than horizontal may prove more efficient at radiating a signal.

8.1.4 Path Loss Path loss (sometimes called free space path loss) is a degradation of a wireless signal as it travels. Like all radio waves, distance often hinders the ability to keep the signal clarity at a high level. As a radio wave travels, its waveform slowly starts to deteriorate with distance. This is because path loss is occurring. Path loss is often important for bridge solutions, because they extend long distances. It is typical that line of sight stays within six to eight miles in bridging solutions to prevent path loss. If this is not followed, the path loss of the signal might prevent it from being received on the other end. Figure 8.2 reveals path loss: as the antenna signal travels, it gets degraded until it is so minute that it cannot be detected. This can be compared to the discussion in Chapter 1 relative to dropping a stone into a pool of water and watching the ripples until they cannot be seen.

8.1.5 Azimuth The azimuth (Figure 8.3) of an object is the angular distance along the horizon to the location of the object. Azimuth can be used to apply direction toward the flow of a current. This current flow usually reads as a degree notation from the north point. When using azimuth with respect to wireless, its main function is to aid in the placement of antennas. We can illustrate, in a design, which way the wireless signal is propagating from the antenna by relating its azimuth. This helps orient the wireless signal’s intended radiation pattern on a two-dimensional drawing.




115

Azimuth 0° N 0°

Azimuth 90°

W 270°

E 90°

S 180°

Figure 8.3 Antenna azimuth.

Pa th

on e

Mutipathing

o tw th Pa

Figure 8.4 Multipath.

8.1.6 Multipath Often, a signal will perform one of the interference types, such as reflection or diffraction. When this happens, sometimes the signal splits into multiple signals (Figure 8.4). When more than one of these signals is able to make it to the receiver, they can cause a number of problems. The reflected RF waves can travel farther than the original RF wave, thus arriving later in time than the direct RF wave. This can make it extremely difficult for the receiving antenna to put the signal back together. Another effect of multipath is distortion, which can come from having a wave that split © 2006 by Taylor & Francis Group, LLC


116


Antenna diversity AP sees two signals and picks the best one.

th Pa

e on

wo

t th Pa

Figure 8.5 Antenna diversity.

during travel, creating a different wave that ended up at the receiver at the same time. When one of these events happens, it is called multipathing. This can cause issues with RF transmissions. When RF uses modulation techniques to enhance its speed, symbols are used. These symbols can become corrupt because of multipathing. When this happens, the transmitter must be able to detect that the signal is corrupted and resend the signal.

8.1.7 Antenna Diversity To prevent multipathing, an antenna method called diversity (Figure 8.5) is often used. Diversity is the use of two antennas for each radio. This can lower the effect of multipathing by leveraging two antennas to receive a better signal. It does this by selecting which antenna received the best signal. This means that only one antenna is used at a time. Diversity antennas are physically separated from the radio and each other, allowing them to be placed in areas outside the propagation of multipath. Using dual antennas ensures that if one antenna is receiving a high amount of multipath and the other is not, diversity allows the radio to pick the antenna with the better-received signal. When looking to maximize the advantage of diverse antennas, make sure that they are placed two to eight feet apart. They should also be the same antenna type covering the same area. Often, people use a diversity-capable access point to cover two separate areas. This can increase the effects of interference. Also, placing antennas less than two feet apart may still help, although it will not provide maximum protection from multipathing. © 2006 by Taylor & Francis Group, LLC



117

Fresnel zone

Figure 8.6 Fresnel zone.

8.2 Fresnel Zone The Fresnel zone is the area around a wireless signal that needs to be clear of any obstructions in order for a wireless signal to be accurately received. The wireless signal itself is not a small airwave; it has a large breadth and objects can often get in the way. When looking for a line of sight, which is needed for wireless bridges, one must take into account the Fresnel zone. Looking at Figure 8.6 you can see the fresnel zone affects long range communication. As one can see from the figure an object can interference the fresnel zone and this can create communications issues while still having good line of sight. To fix this issue, a large tower is needed to get the antenna higher than this Fresnel zone. The Fresnel zone is also impacted by the curvature of the Earth. As Christopher Columbus discovered, the Earth is round. How he proved this teaches us about how a wireless signal could become blocked simply by the curvature of the Earth. When Columbus told everyone that the world was round, the public asked him to prove it. When he left on a short trip out to sea, he asked everyone to watch his sail as he traveled. As he went out to sea, his mast started to slip out of sight. As the people on shore watched, right before his ship went totally out of sight, all they could see was the tip of his sail. Columbus said, “If the world was flat you would be able to see the entire ship”; but because the world is round, the body of his ship slipped below the curvature of the Earth. Funny enough, no one really bought this theory and he still attempted to travel around the world to prove his point. © 2006 by Taylor & Francis Group, LLC


118


8.3 Antenna Types Given a basic understanding of how antennas are measured and what variables affect their usage, one can start to look at the different types of antennas and how each antenna type propagates signals in a different patterns. Given a full understanding of the different types of antennas, one can then look at some of the primary connectors used. Finally, there are homemade antennas similar to the ones primarily used by war drivers.

8.3.1 Directional Antennas A directional antenna has a radiation pattern crafted to flow in a single direction. This single direction flow is very difficult to accomplish because all directional antennas will radiate a small amount of signal 360° around the antenna itself. If one were to look at how a directional antenna is created, one would see that when one squeezes an isotropic radiator, one pushes the signal out one direction while some of the signal squeezes out the other end. This means that if one is standing directly behind any directional antenna, one will see that the signal still is able to radiate the opposite way of its intended direction. When looking at antenna placement, directional antennas are best used to shoot signals down long narrow paths such as hallways and aisles. They are also good for shortrange bridging applications. Visit a large warehouse such as Home Depot and one can see the usage of directional antennas. The most common directional antenna is the Yagi. This antenna generally has a 17° beamwidth, closely representing a triangular cone shape. In general, the parabolic dish is specifically used for long-range bridging applications. Its beamwidth and rotation pattern are very similar to that of the Yagi except the signal will travel much farther. The patch antenna is more of a middle ground between directional and omnidirectional, as its beamwidth is 180 °. The patch antenna is often used against walls because of its 180° coverage pattern. It is still classified as a directional antenna. Another reason the patch is often put against walls is to prevent the wireless signal from leaking outside. Figure 8.7 shows both Yagi and patch antennas.

8.3.2 Omni-Directional Antennas Looking at other forms of antenna, there are the omni-directional antennas (Figure 8.8). What differentiates omni-directional and directional antennas are the beamwidths or coverage patterns. All omni-directional antennas have a 360° coverage pattern. This means that they radiate equally in all directions. Omni-directional antennas are best suited for large, open areas © 2006 by Taylor & Francis Group, LLC


Wireless Antenna Theory 119

Figure 8.7 Directional Yagi and patch antennas.

Figure 8.8 Omni-directional antenna. © 2006 by Taylor & Francis Group, LLC


120


that need coverage. Placing omni-directional antennas against any walls or directly on a column will prevent their ability to properly radiate the signal. Often, omni-directional antennas are hung up against columns; this is due to their usage in large, open areas such as warehouses. When this happens, it is very common for someone to mount the antennas right against the column and not place them out from the column where they would operate more efficiently. To get the best coverage, one should place omni antennas at least three feet from any column.

8.3.3 Homemade Antennas Homemade antennas (Figure 8.9) have become popular due to the large price tag on commercial antennas. A Yagi antenna, similar to the one constructed out of a simple Pringles can, will sell for almost $200. There

Figure 8.9 Homemade antenna. © 2006 by Taylor & Francis Group, LLC



121

are even some Web sites detailing how to make a Pringles can antenna — they say the cost is about $6.50. One thing to note here is that the FCC considers homemade antennas illegal. There are some homemade antennas for sale. Outside the United States where the FCC has no authority, one can purchase homemade antennas. One place in Canada operating an online shop (http://www.cantenna.com) is selling homemade Pringles can antennas.

8.4 Connectors Connectors are used to hold the antennas to devices. They can be used to hold the antennas on access points or wireless network adapters. There is not much available information associated with antenna connectors. However, one needs to ensure that one uses or orders the same type of connectors or the antenna will not fit. Wi-Fi antennas are very different between access point manufacturers. This means that access point vendors need their own connector types. This section identifies the most common connector types, looking at what access points or wireless network cards use each type. The reasoning behind why there are these defined antenna types goes back to the governing body, the FCC. The FCC wanted to prevent people from using common high-gain antennas and affixing them to wireless access points, thus breaking their predefined power output. This is why they made the industry use those antenna types that are not common across other types of antennas. This is referred to as the “use something uncommon and you will be OK” rule.

8.4.1 N Connectors The N connector was created by Paul Neill of Bell Labs in the 1940s. It was named after Neill and call N for short. The N connector is common among antennas; most antennas come with N connectors. This is a little different from Wi-Fi connectors. Almost no access points use N connectors to attach to antennas directly. When ordering antennas, some only come with the N connector and a converter is needed. Knowing what the N connector looks like will help prevent problems. Figure 8.10 shows what an N connector looks like.

8.4.2 Reverse-Polarity TNC-Type Connector (RP-TNC) The reverse-polarity TNC-type connectors (Figure 8.11) are one of the most common access point connectors. This connector, like the N con-



122


Figure 8.10 N connector.

Figure 8.11 Reverse-polarity TNC-type connectors.

nector, was created by Paul Neill, although in this case another person also contributed, giving us the Concelman part of the name. His name is Carl Concelman and he worked for Amphenol. In the late 1950s, Neill and Concelman created the TNC connector to reduce electrical noise, which was causing problems in the common BNC-type connectors. This noise was only seen under extreme vibrations. The reverse polarity part © 2006 by Taylor & Francis Group, LLC



123

(RP) is created by a keying system with a reverse interface. This ensures that reverse polarity interface connectors do not mate with standard interface connectors. Concelman and Neill accomplished this by inserting female contacts into plugs and male contacts into the jacks. Some manufacturers might use reverse threading to accomplish reverse polarity. Either way it is done, it meets the mandate set forth by the FCC that each wireless LAN antenna type be an uncommon one. To meet this criterion, many manufacturers used this antenna and reversed the polarity to create the RP-TNC connector. A large amount of access points, including Cisco, Linksys, Proxim, and Sonicwall, use the RP-TNC connector. For the most part, the access point is the female side of the connector. This means that the antennas connecting to these units must have male RP-TNC connectors. Most antennas sold solely for 2.4-GHz wireless systems have male RP-TNC connectors. If the connector is an N connector as described above, then a pigtail converter is needed. This allows the RP-TNC connector to work with an antenna with an N connector.

8.4.3 SMA, RP-SMA, and RSMA All of these connectors connect a client card to an antenna. In SMA (subminiature version A) connectors, as for RP-TNC, each manufacturer met the odd requirement by using reverse polarity, thus giving the RP portion. The SMA connectors were created in the 1960s. The connectors in this section primarily are used for Peripheral Component Interconnect (PCI) wireless cards. This means that they are used in desktop computers, as compared to Personal Computer Memory Card International Association (PCMCIA) cards, which work inside laptops. The SMA, RP-SMA, and RSMA connectors are used to connect the wireless adapter to an external antenna. When a wireless card is used on a desktop, the placement of the tower and the antenna is most likely behind the unit under a desk. This is not the best place to put an antenna; because of this, these connectors allow for a variety of antenna types and the ability to place them on top of the desk where they are more likely to get better signal levels.

8.4.4 MC and MMX These types of antenna connectors are mostly found in PCMCIA wireless cards. MMX is an acronym for “micro miniature”; these connectors are called that because of their small size. The MMX connector type was created in 1990. Most wireless cards fall into this category. Usually, these connectors are placed with a fixed antenna on the card, making the cable



124


Figure 8.12 MC and MMX.

never seen. Some cards allow for antenna modifications. To allow for adding an extra antenna, they either offer no antenna and one that allows for the connectors, or a fixed antenna with one of these connectors that bypass a fixed antenna. Looking at Figure 8.12, one can see that one of the cards has no antenna and just the connector, and the other one has the connector at the end of an antenna. On the second one, when an antenna is hooked up to it, it will bypass the fixed antenna and use the external antenna. Knowing which one of these connectors is on a radio card will ensure that the correct antenna is ordered. Figure 8.12 shows these types of connectors on a couple of PCMCIA wireless network cards. These connectors are some of the most common types of pigtails used for war driving. When someone wants to war drive, he or she will most likely find a way to use powerful antennas on their client cards. To note is that sometimes, overpowering antennas like the Yagi can actually reduce the strength of an antenna when connecting to a client card. This is due to the small amount of power outputted from these cards compared to the large amount of power needed for a large Yagi antenna.




125

8.5 Chapter 8 Review Questions 1. When radiating an RF signal to create an area of coverage, what part of the WLAN is used? a. Client card b. Access point c. Antenna d. None of the above 2. What happens to the radiation pattern of an antenna when the gain is increased? a. The angle of radiation becomes larger. b. The angle of radiation becomes smaller. c. Nothing. d. The coverage decreases. 3. Which antenna allows for coverage in a 180° area? a. Yagi b. Patch c. Omni-directional d. Directional 4. What happens when a 50-foot antenna cable must be replaced with a 100-foot cable? a. Nothing. b. The area of coverage will decrease. c. The area of coverage will increase. d. The antenna cable will reduce the access point’s transmission rate. 5. Which antenna allows for coverage in a 360° area? a. Yagi b. Patch c. Omni d. Directional 6. EIRP is a representation of the power that an entire RF system has. When correctly measuring EIRP, what should be measured? a. Cabling output b. Radio output c. Antenna output d. Client card antenna input



126


7. What connector is used on a 2.4-GHz 1200 Series Cisco access point? a. RP-SMA b. SMA c. RP-TNC d. MC-MMX 8. How does one calculate EIRP? a. Antenna gain + Antenna loss b. Antenna gain + Cable loss − Transmitter power c. Transmitter power + Antenna gain − Cable loss d. Cable loss − Antenna gain + Transmitter power 9. To help with null areas and multipathing, what antenna architecture would you recommend? a. Splitters b. High gain antennas c. Low loss cables d. Diversity 10. Some access points have two antennas on them. Why? (Select two) a. To increase the speed of the wireless b. To reduce the effects of multipathing c. To provide a hot standby antenna d. To provide antenna diversity 11. There is a customer who wants to connect one office building to four warehouses in the area. Most of the offices are located less than a block away; the farthest one is half a mile away. Placing what antenna on the roof of the office will provide the best coverage? a. 13.5-dBi Yagi b. 6.5-dBi patch c. 21.5-dBi dish d. 5.2-dBi omni 12. What is the designation for dBi? (Select two) a. Dipole b. Decibel c. Baud d. Isotropic




127

13. The area around or immediately surrounding a wireless signal between two locations is known as the ________________. a. Free space loss b. Line of sight c. Fresnel zone d. Signal area 14. What is the typical line of sight when using wireless between two buildings? a. 2 miles b. 11 miles c. 15 miles d. 6 miles 15. You are trying to provide coverage down a long hallway. What is the best antenna solution for this? a. Yagi b. Patch c. Omni d. Directional 16. Determine the EIRP for the following configuration: a 20-dBm radio using a 13.5 Yagi antenna with a 100-foot cable at a loss of 8 dBi per 50 feet. a. 17.5 b. 254 c. 33.5 d. 29.5 17. You are connecting two buildings together with a wireless bridge. You are using a Cisco bridge with a Yagi antenna. In between the two buildings are some trees. You are unable to correctly set up a connection. What should you do? a. Call Cisco TAC. b. Use a dish antenna. c. Raise the antenna above the treetops. d. Use a high power setting. 18. If an antenna has a 6-dBd rating, what is that rating in dBi? a. 8.12 b. 9.31 c. 4.8 d. 8.14 © 2006 by Taylor & Francis Group, LLC


128


19. When an antenna gain increases, its beamwidth _______________. a. Increases b. Decreases c. Stays the same d. Multiplies 20. Which antenna is the most directional? a. Yagi b. Patch c. Omni d. Parabolic dish 21. What is the designator for dBd? a. Dipole b. Decibel c. Diversity d. Baud



Chapter 9

The Wireless Deployment Process The process of installing a wireless network is often simple for a single deployment. When looking at a controlled process that is repeatable across the entire enterprise, there are number of steps that one must address. This chapter explores each of these steps, looking at ways to implement controls that help lower costs and create consistency across the enterprise. When deploying wireless networks on a large scale, one must look at the most controlled, cost-effective, and repeatable process available. Getting to this requires some up-front work and a lot of experience. When deciding to take on a large wireless project that spans the enterprise, one of the first decisions is whether to perform the work inhouse or outsource it. Often, using an outsourced model is one of the better ways of tackling a wireless deployment. When vendors are involved, some of the risk is shifted from the organization to their vendors themselves. Ultimately, the company is still responsible no matter what. If this process is properly handled, the vendors assume any additional financial risk outside the contracted amount. Often, unforeseen financial burdens will arise, stemming from a poor design, unskilled designer, malfunctioning hardware, or any other technical hurdle that might impede the timing or cost of a project. This approach is more difficult, if not impossible, to apply to smaller organizations; however, for larger companies, this route performs especially well. When upper management is more financially driven than technically driven, having the ability to fix the cost associated



130


to any technological deployment will greatly improve your worth as an employee. Looking at the entire deployment process, we can walk through a fictional organization (MoreCo Corporation), addressing its requirements to deploy wireless. The wireless system will be installed to support the organization’s physical security force. MoreCo needs to deploy wireless at 20 locations nationwide. In starting this process, one gets into the requirements for wireless, looking at the need for wireless. This section addresses the why and what the organization can expect to gain from this technology. This will help ensure that at the end of the day, the project requirements allow the customer to achieve what they originally needed. In walking through the entire process, one will see how MoreCo handled each step of the process. This information will help by providing a more true-to-life example of how to perform the steps outlined in this chapter. If a deployment falls within the size scope described above, a key element would be to have an independent, technically qualified consultant available. This consultant might be an internal employee. If an internal employee is used for this role, make sure he or she is technically qualified and has sufficient experience within the wireless industry. One of the best approaches is to hire a short-term consultant for the duration of the deployment. The focus of this consultant would be to oversee all technical aspects of the process. Having this resource involved can bring valuable insight to your team and can help develop and implement a number of controls identifying any technical shortcomings of the various vendors. By using this resource in an independent auditing role, one can often find several issues that otherwise might not be brought to anyone’s attention until it is too late. Having this kind of expert analyzing all of the designs and installations forces the vendors to optimize designs and install the most technically applicable wireless networks possible.

9.1 Gather Requirements First, one must look at exactly what a company will use its wir eless network to accomplish. This is one of the simplest tasks for many people because the greatest benefit of wireless is the increase in worker productivity. There are many other applications for wireless across many vertical markets, such as retail, manufacturing, education, healthcare, and many others. In these industries, one often sees wireless being used as a tracking method to reduce the overhead and labor costs associated with inventory. In addition, one sees wireless point-of-sale systems appearing in the retail



The Wireless Deployment Process

131

market, allowing easily movable storefronts. In healthcare, one sees doctors and nurses using wireless tablets, creating a virtual clipboard with all the hospital records on it. There are a number of reasons in each vertical market for deploying wireless. The main goal here is to identify the driving requirement for wireless. After determining what the driving requirement or requirements are, the next step would be to find out how much wireless is needed and where. What this means is, do you need wireless everywhere within your facility or maybe just a small area? Understanding the reasoning behind deploying wireless gives a good idea of where within a facility or campus it is needed. In a wireless deployment, there are often many people involved with many different skill sets. Gathering requirements from each group, as well as understanding them, is a key role in the requirements stage of a project. A person who is not technical may have requirements that unbeknownst to them may need technical solutions or may change the general requirements altogether. This is often seen when someone requests wireless access inside places like a cooler. It is possible to place wireless inside a cooler, although it will change the pricing. This pricing change may be substantial enough to drop that requirement altogether. In applying this section to the example company MoreCo, one can see that the businesses need is to wirelessly hook the physical security force into the camera system from their PDA devices. This means they need wireless in all the areas of the facilities that the security force are required to guard. This would be a prime case where understanding the skill sets of individual job roles would help to make sure that the most cost-effective view is taken. At MoreCo, there is a shed at one location; this shed is more than 2000 feet from any telecommunications closet. To add to this, there is no line of sight to use a bridge solution. What this means is that if wireless is needed inside the shed, a major cost increase will occur because fiber optic cabling will be needed to support that single access point. This cost increase may be fine with the technical team leaders; however, the managerial team leaders may decide that the cost is too great and change the requirements away from covering all buildings. The next requirement to address is the bandwidth required to support video. Video is real-time, which means it needs bandwidth. This means that certain data rates must be everywhere there is coverage. This is needed as a requirement so that when the wireless network is put in, it can support the high data rates required to stream video. Having dealt with the main requirements issues, one now understands why wireless is needed and where exactly it is required. One also knows that the wireless network needs to support time-critical data exchange stemming from video. Other key points that one must address include when it is needed, how much it will cost, and how it should be done.



132


9.2 Estimation With a good understanding of what a wireless network will be used for and what the requirements are, one can then look at one of the most significant factors: the cost. One needs to gather this prudent data to move on to the next step, that of making a business case. In looking at the cost of wireless, many factors create a fluctuating cost. Even in buildings that are designed by the same architect and have exactly the same layout, one will see a difference in RF design. This is due to the many environmental variables that are outside one’s control. These variables often influence wireless coverage patterns, changing the amount and placement of equipment. The estimation stage is listed second in the process. This is done for a two main reasons. First, to get an accurate cost of a wireless deployment, one must perform a site survey. This site survey takes time and resources to perform and, because of this, the survey often has a cost associated with it. Second, if one were to make a business case before actually getting an estimated cost, one could not properly report one of the biggest factors needed in the decision-making process. Therefore, one needs to have a sound way to estimate cost without performing any cost-prohibitive tasks. The estimation process of a wireless network can be a tricky one for a number of reasons. The best way to approach this is to use an average area of coverage based on a single access point and then divide that number by the total square footage of the area where coverage is required. Keep in mind that this will change, depending on the technology used (802.11b/a/g). This is because each technology has its own ar ea of coverage size. Right here would be one of the first areas that independent consultants could come into play to provide personal advice and experience on the subject matter. They will know, based on the requirements, what kind of technology will best suit everyone’s needs. This will give a rough estimation of cost. Although it will not be 100 percent accurate, it should prove close enough to make the business case and receive the funding needed to proceed to the next step. Many other costs must also come into effect during this stage. The wired network must be in place and able to handle the access points. There might be cable length issues and other issues that may incur extra costs (e.g., outdated equipment). Most of this detail would not be discovered until the designing phase. If the correct questions are not posed upfront during the estimation stage, the cost could change dramatically. This would be another reason for having an expert on your team who knows what has burned them before and now knows what to avoid. MoreCo made a choice to use the 802.11g standard. They made this choice based on testing a small consecration of access points from the vendor that they wanted to use. This testing proved that their application functioned correctly and handled all the requirements correctly. They then © 2006 by Taylor & Francis Group, LLC



133

took this number and applied it to the formula above to estimate the number of access points they might need. Afterward, they sent checklists to all 20 sites. These checklists were created by the independent consultant, outlining questions that if answered incorrectly would raise red flags. One example would be: “Do you have locations requiring wireless coverage that are more than 300 feet from a telecommunications closet?” Once these were sent and returned, an estimate can be created. This estimate provides a number close to what the project will actually cost.

9.3 Make the Business Case This section looks at what is required to convince management that a wireless network is necessary. Using the increase in productivity is often very helpful in this area. When looking at the other vertical markets, inventory tracking, material handling, and general mobility often make good business cases of their own. In creating this business case, using real-world statistics from reputable marketing firms will reinforce one’s business case. If using a wireless network to increase productivity, try using some of the research done on the amount of increased productivity due to a wireless LAN. This has been researched and documented by a number of marketing firms. One of the key points in making a business case is return on investment (ROI). How can this wireless system save the organization money? Is it by reduced costs for material handling or labor associated with inventory tracking? Is it increasing productivity because team members will be connected no matter where they are inside their office or campus environment? There are many benefits to wireless networking; to make it easy to quantify cost savings into ROI, one must understand the benefits from a cost-savings or profit-producing prospective. This is one of the more challenging tasks. Once one creates numbers based on anticipated savings or anticipated productivity gains, those numbers easily become speculation. As pointed out above, the best way to approach this is to use industry marketing information as the basis for the amount of increased productivity. Most management teams that do are only interested in technology that can improve their bottom line and not introduce excessive costs to support and maintain. If one can present a wireless network that saves money and is backed up with statistics from other companies, then one’s business case will soon become a funded project. When looking at the business case from the standpoint of MoreCo, one can see that the benefits include security and cost reduction. Because of these benefits, the new wireless system will improve incident response time. MoreCo will see a benefit of better security because their physical



134


security team members will be in constant contact with all available cameras and command centers. One can see the benefit of cost reduction because now when team members respond to an incident, they are not required to have someone watching the camera and reporting what they see. With the new solution, the responder can get updates from his PDA en route. This solution saves money associated with having two security personnel responding to each dispatch.

9.4 Site Survey Once management has signed off and funding has been acquired, the first step of the physical work takes place. The site survey is used not only to get an accurate cost, but also to get the correct data needed to design the wireless network properly. When performing a site survey, there are a couple of requirements. The first requirement is equipment. The equipment needed to perform a site survey consists of what is called a site survey kit. Inside this site survey kit is an access point, a large array of antennas, a portable battery pack, and a large vertical boom used for raising the antennas. This equipment is used to try a number of different antennas to achieve maximum coverage within any given ar ea. The surveyor will need architectural drawings showing the layout of a building or campus. The surveyor can create these drawings, although it is more cost-effective to provide them. The site survey team also needs to have a spectrum analyzer so they can locate and identify any inter ference. Depending on whether a bridge solution or a large outdoor solution is needed, the site survey vendor might need a GPS device. This section discusses how to perform a site survey and how to properly manage it to the customer’s advantage. Many companies perform site surveys. Choosing the correct company requires a little background work and proper vendor management. The first process is to obtain a qualified and reputable site survey vendor. To do this, one would need to release a Request for Proposals (RFP) or a Request for Quotes (RFQ). To maximize the success level, there should be a number of vendor requirements inside this document. These requirements are one area of the process where one can implement some technical as well as financial controls to mitigate risk.

9.4.1 Performing the Site Survey When performing a site survey, there are a number of items to address. Before one can perform the survey, one needs to make sure that one has the correct equipment. Bringing the proper equipment will ensure that © 2006 by Taylor & Francis Group, LLC



135

nothing is missed and that all the surveys are performed correctly. One thing to note about equipment: if performing any type of work far away from one’s office, always assume that everything will break. With this said, if you are performing the survey and something breaks, it is your fault for not assuming that everything will break. If managing a wireless project and a team member or vendor says that something broke and it prohibits them from working, make them aware that it is their fault. They should have assumed that everything will break, and not doing so is a lack of experience on their part. Many companies will sell full-blown kits that have most of the items discussed above. In every kit, there will most likely be something that is needed or not provided. This subsection looks at each of the needed items, identifying why one needs them and how they work. Then it examines the process of performing the survey. The first item one needs is an access point. This access point must be the same model that is going to be installed after the survey. Using a different access point model for a survey from the one that will be installed can and often does create coverage problems. After performing the steps outlined in the gather requirements step, one should choose a wireless transmission technology. Now all that one needs is direction on the access point manufacturer. Currently, there are several manufacturers. Making a choice between them depends on a number of factors, such as current relationships, industry leaders, special requirements, features, or any other factor. Some of the more dominant access point vendors include:

Cisco Aruba Proxim Symbol SonicWall Netgear Linksys D-Link

The next item is a method to power the access point. Providing power to the unit is critical. Not only does it need power to operate, but it also needs power without dragging cords around the building being surveyed. In most cases, there will not be a plug available to use where one wants to place the access point. To fix this, one can purchase a battery to provide the access point power. Depending on the length of the survey, one battery can last for the duration of the survey (two batteries may be required). One should always come prepared, always come expecting everything to break. Some companies sell these batteries as part of their © 2006 by Taylor & Francis Group, LLC


136


surveying kits. Use the Internet or access point manufacturer to locate companies that sell site survey kits. The next item is a telescoping pole, which is needed to hold together the access point, antennas, and cabling. This pole must have the ability to be raised and lowered to accommodate the height at which the access point will be mounted. Normally, this pole is constructed or purchased with the surveying kit but some kits do not provide the pole. The pole should be light and telescopic so it can be broken down to be shipped or put on a plane. The pole should be a bright color to prevent people from running into it. This may sound like a small issue that can be skipped — until a surveying team is told it cannot erect the pole due to it not being a highly visible color. Many safety regulations require this, so make sure this does not become a showstopper. To perform a survey, an array of antennas should be part of the surveying kit. These antennas allow the surveyor to have maximum flexibility when placing access points. There are many types of antennas for access points. Using all of them is more of a hassle than a benefit; not only does this become a problem for the surveyor to lug around, but it also becomes a problem in supplying spares. When a network has nine or ten different antenna types, keeping all the spares can waste some money, especially if two of the antennas are very similar. The best way to handle this is to standardize on a couple of antenna types. To figure out which ones to pick and which ones not to pick, one should look closely at which antennas are the same type and close in power rating. Choosing two patch antennas, one at 5.5 dBi and one at 5.2 dBi, is a clearcut example. Having the same type of antenna with less than 0.3 dBi difference is a waste of money and time. Also, when choosing to limit the allowable antennas, make sure to include, at a minimum, one of each type. An example of this would be to choose one patch, one omni, one Yagi, and ceiling-mount variation. This will ensure that one is providing maximum flexibility without increasing complexity. Another, hopefully obvious item is a wireless end device. Most likely, this will be a laptop with a wireless card. The wireless card should be the same model that will be used in the end clients. This is not always possible, although it is recommended. In addition, when looking to pick a wireless card for surveying, make sure to not have an extremely highor low-powered card. Using an average card will ensure that the survey is as accurate as possible. Two of the same cards from the same vendor can often have very different signal results. To make sure the survey is as accurate as possible, use a number of cards beforehand to find the most common one. Having found a common card not too strong and not too weak, mark that as a surveying card and use it. If one uses this same card for awhile, perform the testing again and always make sure that at




137

least 25 percent of the cards used for testing are ones that have not been tested previously. When making an initial surveying card purchase, make sure not to buy the cards together. The differencing factors of client cards can rarely be seen from cards that were part of the same factory run. Some other key items to bring include the tools required to document the results. These tools include a set of drawings, colored pencils, bright tape, and some colored sticky dots. The pencils and dots are used to draw the coverage patterns and corresponding access points on the drawings. The bright tape is used to indicate the actual area where the access point should be placed. The tape is affixed to the column in the exact place that the access point will be hung. The final item is a spectrum analyzer, which is used to find any interfering signals that might impede a wireless network. Because wireless local area networks work in unlicensed airspace, other devices might also be using the same frequency. In addition, other access points might interfere with the survey. These access points could be from other, nearby companies or employees who could not wait for wireless. One of the reasons for the analysis is to help the customer and the surveyor know that the air at the frequency which the wireless network is operating in is clean. Many types of spectrum analyzers are available, ranging from inexpensive $1000 models to high-end, $100,000 models. To perform a simple analysis to see if another signal exists is easily done with the least expensive models. To use the analyzer to troubleshoot problems, a highend analyzer is a more appropriate choice. Note that the higher-priced analyzers are available to rent. This means that in the rare event that one might need one, there is no need to worry about shelling out $100,000. To perform the survey, a team usually consisting of two or more people will sit down and get a good idea of where and how they want to cover a given area. All too often, the surveying team goes out and just starts putting up the pole and walking off the coverage circles without sitting down and planning the logical placement of the access points. When this happens, there is often a waste of coverage and a crude-looking design. The correct way to perform a survey is to perform a draft run on paper, looking at the most feasible antenna options for particular areas. Understanding this and following it will improve time and reduce the cost. Once this is completed, the survey physically starts. Now the team goes out to the first location and raises the telescoping boom with the access point and the first antenna candidate. Once this is up, they connect to it with the laptop and make sure that the signal looks good and the device is sending and receiving traffic. Next, they walk off the area of coverage. This is done by locating the edge of the signal all the way around the access point. There are many tools available to do this. Chapter 1 discussed SNR values; these values play a key role in



138


determining the end of the cell’s coverage. Once the end is found all the way around an access point, it is detailed on a drawing with a colored pencil and a dot is placed where the access point should go. The next step is to mark, using the bright tape, the location of where the access point should hang. Affix the bright tape to the column or something that represents where the access point will be mounted. After this, the team moves to the next location and repeats this process. The entire area where the customer asked for coverage goes through this process. After performing the survey, the drawings must be converted to an electronic format. The most common format is auto-CAD. During this process to convert the documentation, channel selection must be addressed and designed. When performing the survey, looking at a channel structure is omitted. One is not required to pick channels during the site survey. Performing a spectrum analysis makes sure that no interference is inherent on any of the available channels. Because of this, it is common for a site survey designer to develop a channel structure when all the access points are listed on the drawings. The next set of documentation involves a physical information sheet showing all the access point locations. Inside the physical information document are the following items:

Location of access point Placement of antenna Type of antenna Power settings Channel settings

Going deeper into the physical information document, take a look at each of these items, what they are, and why they are needed. The location portion documents the bay and column where the access point is supposed to be mounted. This helps whoever might install the access point find the correct location. During an access point installation, placement of the access points is very critical. Moving the access point more than seven feet from where it was surveyed can dramatically affect its coverage pattern. This is why location information is required in the site survey design. The next two items, antenna placement and antenna type, are addressed together. These are used to detail and explain how and what antennas are mounted and in what locations. This is important because certain antennas must be mounted in certain ways. Mounting a patch antenna pointing the wrong way will quickly become an issue. With correct details, this can be avoided. Having this item in the documentation ensures that events like this never happen. This also helps against installing the incorrect antenna altogether. © 2006 by Taylor & Francis Group, LLC



139

The next portion is very important; it details the access point’s channel and power settings. The power settings dictate how big a coverage pattern is. Most commonly, a site survey vendor will blast these devices with full power. This is not a recommended practice because one does not have the ability to increase the power. The best practice approach is to inform the site surveyor that they cannot use the maximum or minimum power settings. This allows for flexibility if a coverage issue arises in the future.

9.4.2 Technical Controls As discussed in Section 9.4, there are some controls and requirements that can be put in place. When using a company to perform the survey, a number of requirements and controls must be put in place well before any work starts. If these controls are thought out and implemented in an early stage, the risk involved will be dramatically reduced. One of the first controls would be to ensure technical competency. Having a statement requiring the vendor to have trained and certified individuals on staff is a start. All too many times, companies win bids for project they subcontract out. When they do this, they do not require the same technical competencies originally asked of them. To make matters worse, depending on how it is worded, they may have fulfi lled the requirements just by employing technically competent resources. This does not mean they will work on your project; be careful of this, as it happens all the time. Making sure to stipulate that certified individuals are working on your project is one of those small details that can create big problems later if not addressed up front.

9.4.3 Financial Controls Looking at a financial control, first make sure that the requirements for the vendor are clearly defined and specifically detailed. This must be complete and accurate if this control is to work correctly. Understanding exactly where and what must be covered by a wireless signal is necessary for this control to work. If any of this information is incorrect, the control will lose its power and become meaningless. If all this is done correctly, the agreement between customer and site survey vendor can hold the site survey vendor responsible for ensuring that the survey will adequately cover the required area until certification. Providing warranties for a site survey has long been a gray area for many site survey vendors. This is because environmental variables can adversely affect a site survey. This means that if there is no interference when a survey is performed and then suddenly there is, who is responsible? This again is when and where an independent party can come into © 2006 by Taylor & Francis Group, LLC


140


play, providing positive feedback from previous experiences. Most site survey vendors will only guarantee their surveys for a very limited time. Every day the site survey is under warranty past the day the survey was performed becomes additional risk for the site survey vendor. This is why site survey vendors have such short warranties. Correctly weighing this risk versus the project timeline is a tricky task that must be adequately understood by all involved parties. Correctly documenting the length of time and full extent of contractual obligations to warranty a site survey is necessary before any work begins. If this step is done correctly, the site survey contractor is fully responsible for the financial cost of designing, installing, and certifying any additional access points if the required area was not fully covered. This can also go the other way with too much coverage; too much coverage will cause just as bad of problems as not having enough. Having too much coverage is extremely difficult to control from a contractual standpoint. Having the consultant make his recommendation is key at this point. If the survey looks excessive in relation to the total number of access points, the consultant can easily point that out. When we look at MoreCo specifically, their best option would be to have the independent party write up a Request of Quote (RFQ) document. This RFQ should detail the length of warranty and the extent of what financial responsibility the vendor would have in the event of a coverage issue while under warranty. The control on too much coverage is implemented by having the consultant approve the site survey document before it can progress to the next step of design. More then likely, this party would have close interactions with many vendors that perform this service and would already have a good idea of who has successful and unsuccessful surveys under their belt. This could become an added benefit by allowing other specialized companies that the consultant was aware of the opportunity to bid on the project. Later in the certification section we will see how these controls are set up.

9.5 Design In the design stage, one looks at taking the information from the site survey and adding the network logical and physical designs to it. The site survey deliverables for a project include a coverage drawing, access point locations, power settings, channel settings, and antenna usage. There are no logical IP addressing, cable paths, or wired network changes in their package, although most site survey vendors will find the nearest communication closet and give an estimate as to the cable length needed. This is where the design phase comes in; a person or a team will figure out the needed IP addressing and other needed changes for the




141

wired and wireless networks. They will provide drawings on the cable paths and might even provide fiber drawings. They will provide an endto-end view of the network, tying in the new access points and any other devices that might be needed to support the wireless network. When looking at the scenario with MoreCo, the best option is to have the local support staff at each site perform the first attempt at the design. After that has been completed, a review will take place with the site survey vendor, project manager, in-house consultant, and the previous project designer. During this review, a number of other people whose eyes looked over the design will be able to bring a new perspective as well as old lessons learned to the table. This will create a smoother flowing design phase that gets faster and more consistent as it progresses. After the review, the designer will update his package and a final round of review will occur, this time with a limited number of attendees.

9.6 Staging Next comes the staging portion; this can be on-site or at a central controlled location. The equipment needed for a project must be turned on, configured, and tested. This testing is often easier to do with everything set up in one location than trying to go from one end of a building to the other every time you want to test or make a needed change. Once everything has been correctly tested for hardware failures, it can be tested as a whole (or what is called a system test). This test will verify any redundancy built into the design and cause the network to perform as it should in the event of a failure. If the staging takes place in a central location, the added benefit of having the same group or team see all the network designs in relation to a project helps keep consistency across the board. This step will ensure uniformity and repeatability across the enterprise. Looking at MoreCo, the choice was to create a project office where they have a team centrally configure all the new equipment. This allowed them to catch small inconsistencies that create large headaches in the field. This approach also allows for a very well-defined build process, which saves time compared to performing this function at each location with different team members.

9.7 Deployment and Installation Once the equipment is staged, it should be installed. Any new network electronics can be easily racked and stacked. Performing the physical design helps prevent a technician from going to the rack to stack the electronics and finding out there is no room in the rack. Installing the access points can also be tricky. They must be hung in close proximity



142


to the site survey locations. Some site survey vendors will leave a small strip of fluorescent tape identifying where the access point should go. The site survey package should include documented designs showing where and how each access point should be hung. To install them, there are a number of options: perform it in-house; use a contractor; or in some cases, have a union member perform the work. The key is to define how to mount the access points and antennas. Many times, this will create problems because the person doing the mounting has no technical education as to antenna orientation. That person also has no idea of what not to mount them next to. This is where antenna-mounting details come into play; these details must be provided by the site survey vendor or created as part of the physical design. Some installation mistakes are very common. Hanging omni antennas against columns or walls is a grave mistake. Omni antennas should be installed at least three feet away from any column. They should never be installed next to or on a wall. Another big mistake involves antenna orientation. Yagi antennas may need a slight tilt downward; they also need to point in a certain direction to be effective. Patch antennas have similar requirements. They may need to be flipped over; sometimes, they are installed upside down. The reasoning behind this is that they have an 8° tilt toward the antenna leads. Another serious issue surrrounding mounting involves obstructions; when a site survey vendor does its survey, something might not be there, such as a wall or a big metal duct, that is present now. Another mistake that can be prevented is to require a 50’ cable loop in order to easily move an access point if there is a coverage issue. Some people say that this loop can cause induction, although no testing has surfaced proving this. If this concerns a team, they can always install this loop at least 20 feet away from the access point, somewhere along the cable run. In looking at MoreCo, one sees that they chose to have their local site maintenance staff hang the access points. They received all the needed drawings from the site survey vendor and were under the watchful eye of the project manger. This helped ensure that none of the mistakes listed above happened; and with the utilization of the same project manager, lessons learned carried over to the various sites.

9.8 Certification After the system has been fully installed and is online, it must be certified by a third party and the site survey vendor. The third party ensures that the site survey vendor has not overlooked any gaps or possible issues before signing off on it. The third party can be the contracted consultant




143

or even a local site member. This is when the site survey company would end its warranty and lift the financially responsibility for any coverage issues. This is why having a third party perform a certification can help identify any gaps or holes in the coverage area. This is why having a third party perform a certification can help identify any gaps or holes in the coverage area. This is also the time for the site survey vendor to make sure all the access points are installed where wanted, based on their design and documentation. The step is often performed in tandem with many groups that are involved. The site survey vendor, project team, end user, and installation crew are all likely attendees for this step. Not all of these groups or people will need to perform the certification, although if any problems arise, such as incorrect mounting, each team is on-site and ready to make the changes. Next is performing the actual certification. This is often done by following the survey process again, although this time the access points are installed so most of the equipment is not needed. While this is being performed, the site survey vendor is releasing his requirement to fix any coverage issues as each access point is verified to have adequate coverage per the design. Other than coverage, tests are most often performed using the actual wireless clients. This testing ensures that the entire system works and functions at an adequate level while all players are present. At MoreCo, they decided to have their independent consultant perform the third party certification with the site survey vendor. This ensured that the site survey vendor did not perform any work that was not up to the expectations of the consultant, as well as the company. This also gives the site survey vendor a chance to show correct functionality and get a sign-off on their installation. This sign-off will state that the required areas do in fact have adequate coverage. The MoreCo team performed testing of their devices to ensure that the system as a whole operated correctly.

9.9 Audit The audit phase is similar to the certification; although its main objective is not to have everyone sign off on a project as complete. The audit objective is to make sure nothing was done outside of any existing corporate standards and that the network is secure according to corporate security policies. The audit process also ensures that after the network is deployed, it is not changed in any non-standard way. In some remote sites people are often removed from corporate headquarters and are not as forthcoming to ensure compliance. Because of this, security is often pushed away in favor of easier administration or support. This is one of the main reasons why an audit is so critical to the enforcement and assurance that networks are maintained in a compliant fashion. © 2006 by Taylor & Francis Group, LLC


144


9.10 Chapter 9 Review Questions 1. What is the second step in a wireless deployment process? a. Business case b. Design c. Certification d. Estimation 2. When performing a survey, what should the surveyor NOT do? a. Locate any interfering devices b. Correctly place access points c. Provide adequate coverage d. Use local power 3. When performing a survey, what should a survey team always assume? a. More work than expected will be needed b. Everything will break c. Access into a building will be taken care of d. Access point might fail 4. When creating a business case, what piece of data is most important? a. Requirements b. Area of coverage c. Estimated cost d. Time frame 5. When creating an RFQ for a site survey, when should the warranty of the site survey end? a. Business case b. Design c. Certification d. Estimation 6. What is the first step in a wireless deployment process? a. Business case b. Design c. Certification d. Estimation




145

7. What is the final step in a wireless deployment process? a. Business case b. Design c. Audit d. Estimation 8. What item is not normally included in a site survey kit? a. Access point b. Telescoping pole c. Battery to power access point d. GPS device 9. What is a spectrum analyzer used for? a. To find correct coverage patterns b. To locate interference in the 2.4-GHz range c. To find the correct channel to set the access at d. To locate the next location to install an access point 10. Which item inside the physical information document is not correct? a. Location of access point b. Placement of antenna c. IP Address of access point d. Power settings e. Channel settings 11. What step of the deployment process is used to validate the site survey vendor’s coverage and work? a. Business case b. Design c. Certification d. Estimation e. System test



Chapter 10

Wireless Access Points This chapter provides an overview of the many different types of wireless equipment. It predominantly focuses on local area wireless equipment: wireless devices, who makes them, what technologies they support, and in what scenarios they should be used. There are many wireless manufacturers out there, so many that an entire book could be put together to look at each of them in detail. This chapter attempts to address as many as possible, going in-depth only on the most commonly manufactured types. The single most important piece of equipment in a wireless network is the access point. The access point is the piece of equipment that propagates the wireless signal into the air. There are many types of access points from many different vendors. Some of them are made for small office/home office (SOHO) and some are made for large enterprise deployments. Cisco is one of the biggest players in the access point market; others include Proxim, Aruba, Symbol, and SonicWall. As discussed previously, there is an overwhelming number of manufacturers trying to sell access points. Therefore, the focus stays with some of the most used products in the industry.

10.1 Linksys Access Points Linksys was acquired by Cisco in 2004. Thus far, Cisco has decided to keep Linksys separate as a SOHO-only product line. This means that the equipment that Linksys has is primarily targeted at SOHO environments. 147



148


One interesting thing that Linksys has done is move most of its wireless products to an open source Linux kernel. A full list of all the Linksys access points with open source firmware is located below. This has allowed Linksys to utilize some already-existing open source code to achieve lower development costs. Using an open source code has sparked many groups who have hacked into Linksys access points and made some interesting features available within their own firmware code releases. Some of the most common hacked firmware types are Wifibox, Batbox, and Alchemy. These firmware versions are rather easy to install and can quickly improve many of the features of a Linksys access point. Some of them have added functionality that was not present in the Linksys firmware releases. Access points with open source firmware include: WRT54G WRE54G WPG54G WET54G WAP55AG WAP54G Linksys currently has more than 12 access point models. Most of the access points look like the one shown in Figure 10.1. Some newer models have started to take on a more compact design like the one shown in Figure 10.2. This section looks at each of the two styles and sees how they function. Looking at Figure 10.2, one can see the older body style. Linksys has ten access point models that follow this body style. Although this is the older body style, it is more likely that one will run into this access point style before the newer style is out in large numbers. The older Linksys access point has a number of indication lights. Each of these lights allow for an easy glance at what is working and what is not. Figure 10.3 provides an up-close look at these lights. Lights appearing from left to right in Figure 10.3 include: Power light. This light indicator informs the user that power is being received at the access point. This light also signifies that the device is able to start the boot-up process. It will stay on the entire time the unit is powered up to indicate operation. When the unit is undergoing its self-diagnosis testing during boot-up, this light will flash. Wireless light. This light indicates that there is a wireless connection. It will blink if wireless traffic is passing through the access point.



Wireless Access Points

Figure 10.1 Older-style Linksys access point.

Figure 10.2 Newer-style Linksys access point.


149


150


Figure 10.3 Older-style Linksys access point front LED detail.

Figure 10.4 Older-style Linksys access point side port detail.

Port 1–4 light. This light indicates that there is a layer two connection to a wired device on one of the four wired ports of the device. When one of the connected devices starts to communicate, the light will blink. Internet light. This light indicates that an Internet connection is made to the WAN port of the access point. The older-style Linksys access points have many buttons, ports, and connections. Going from left to right in Figure 10.4, one can see all of the items detailing each one. First, on the far left, there is a reset button used to restore the configuration of a device to the factory defaults. Next, there is an Internet port used to attach an Ethernet cable to a high-speed Internet connection such as DSL or cable Internet. This port will allow




151

Figure 10.5 Newer-style Linksys access point front LED detail.

the use of any Ethernet-capable, high-speed access device. The next items are the four wired ports. This group of 10/100 wired ports allow small offices to connect up to four wired devices to the access point. The final item is the power connector. Now on to the newer-style Linksys model. The front panel has some of the same lights shown in Figure 10.3, although some other lights have been added to help improve troubleshooting the device. Figure 10.5 shows the lights on the front of the access point. Lights appearing from left to right in Figure 10.5 include: Power light. This light indicator informs the user that the access point is receiving power. This light also signifies that the device is able to start the boot-up process. It will stay on the entire time the unit is powered up to indicate operation. When the unit is undergoing its self-diagnosis testing during boot-up, this light will flash.



152


Figure 10.6 Newer-style Linksys access point side port detail.

DMZ light. This light indicates the use of a DMZ device. If no device is using the DMZ capability, it will blink during its selfdiagnosis testing, as the unit boots up; once completely up, the unit will stop lighting the DMZ light, unless a device is set up to use the DMZ. Internet light. This light indicates that an Internet connection is made to the WAN port of the access point. Wireless G light. This light indicates that there is a wireless connection. It will blink if wireless traffic is passing through the access point. Port 1–4 lights. This light indicates that there is a layer two connection to a wired device on one of the four wired ports of the device. When one of the connected devices starts to communicate, the light will blink. The back of the newer-style Linksys access point is shown in Figure 10.6. Going from left to right one can see that there is an Internet connection used to access other networks. Next is the four-wired network ports used to connect other wired devices to the network. The first button on the back is the reset button, which is used to reset the configuration to the factory default. The final item is the power connector. Linksys access points use a Web interface to configure them. To access it, a user must type the IP address of the access point into an Internet browser such as Internet Explorer. When the user does this, he is prompted to enter a username and password. On Linksys access points, the default username is admin and the password is admin. Once the user has logged in to the device, a set-up screen appears like the one in Figure 10.7. This will allow the user to set up the access point. This screen allows a user to set up the WAN connection identifying the network information needed to access other networks. It will also allow a user to set up the LAN network information. This device can be a DHCP server, allowing dynamic network configuration information to be pushed out to connecting devices. This screen also allows for a number of advanced features, such as port forwarding, firewall filtering, and MAC-based security.




153

Figure 10.7 Linksys management screen.

10.2 Cisco Access Points Cisco has had a stronghold on the enterprise wireless market since its acquisition of Aironet in 1999. Cisco has created multiple product offerings that fit a large array of needs. Once Aironet and Cisco became the same company, their product line began to move toward Cisco’s own code.



154


Once this happened, the market began to see the major benefit that Cisco’s wireless access points bring to the table. They allow anyone familiar with Cisco router and switch configurations to understand access point configuration with ease. This was due to the fact that Cisco had integrated its IOS code with the Aironet access point, creating an access point that looked, felt, and operated like any other Cisco router or switch. Cisco has three main models of access points: 350, 1200, and 1100 series. Each comes in a wide array of types that use different technologies. For example, the 1200 can use 802.11b, 802.11g, and 802.11a. This section looks at all three types of access points and details what connections they have, what protocols and standards they support, and how they operate.

10.2.1 Cisco Aironet 350 Series The first access point is the older Cisco 350 series lineup. These devices are no longer available through Cisco Systems, although they are still very prevalent across many companies. Currently, the Cisco 350 devices are end of sale. However, they are still not end of life, which means that Cisco will support them. When these devices were sold, they came in a hardened model with a metal casing and in a plastic model that had the antennas affixed to the access point. Both 350 models only work with the 802.11b technology, and that technology only. Figure 10.8 shows the 350 series access point.

Figure 10.8 Cisco 350 model.




155

Figure 10.9 Cisco 350 front LED detail.

Both of the 350 series access points have the same light and connector layout. The only difference between them is the hardened shell and the ability to affix external antennas to the hardened model. To note is that the plastic 350 series access points were available with the option of affixing external antennas, although this was quickly replaced by the hardened cases. Looking at the front of the 350 series access point, one can see in Figure 10.9 that there are three lights. Each of the functions of the lights is detailed below. The lights going from top to bottom in Figure 10.9 are: Ethernet activity light. This light indicates the status of the Ethernet network connection. The indicator will blink green when a packet is received or transmitted over the Ethernet infrastructure. It will blink red when there is no connection to the Ethernet network. Association status light. The association status indicator light signals the operational status of the device. When the light is blinking green, it indicates that the access point is operating nor mally, although it is not connected to any wireless clients. When the light is steady green, it indicates that the access point is associated with at least one wireless client. Radio activity light. The radio indicator blinks green to indicate the presence of radio traffic activity. It is usually off unless there is traffic on the wireless, at which point it will blink green. On the back of the 350 series access point is an RS-232 connection for terminal emulation and a single Ethernet port. On the hardened access point, there are two RP-TNC connectors that attach to external antennas. Figure 10.10 details the connections located on the back of the Cisco 350 series wireless access points.



156


Figure 10.10 Cisco 350 side port detail.

Figure 10.11 Cisco power injector detail.

The 350 series access point has no power plug for the unit. This is because the unit works off Power over Ethernet (POE). POE is a method by which power can travel down a network cable. This power is used to power a small device such as an access point or VoIP phone. To use the POE, one must have a switch that can provide POE. If no such switch is available that can support POE, one can use the small power converter that ships with the access point. This small power cord takes a Category 5 connection from the network and another Category 5 connection from the access point and provides power down the cable. One note about using power injectors is physical real estate. In larger companies, physical rack space is at a premium. Using a power injector means that a physical device like the one in Figure 10.11 needs to exist between the telecommunications closet and the access point. The most logical place to put this device would be inside this closet on a shelf. The physical real estate needed for this shelf and power injector is one of the most commonly missed items when planning for a Cisco wireless network. Looking at Figure 10.11, one can see the power injector that comes with the Cisco 350 access point and one that comes with the 1200 series access point. They have a power connector that is used to connect external power to the injector and two Ethernet connectors (one for a connection




157

to the network, which is not powered, and one connection to the access point, which is powered). The DC voltage used to power the access point is −48 VDC. There is a newly adopted POE standard called 802.3af. Some of the older access points are not considered 802.3af compliant. Cisco states that these devices will work correctly on its POE solutions, although to use another vendor’s 802.3af-compliant switch the access points must be 802.3af compliant. Today, all the access points shipping from Cisco are 802.3af compliant. The operating systems on these units are unlike the familiar IOS that Cisco bases most of its products on. This was due to the acquisition of Aironet; along with acquiring the company, Cisco also acquired the access point code. This meant that the setup and management of the access point is different from most other Cisco devices. To fix that, Cisco needed to change the code into IOS. In late 2003, Cisco created a software code release that converted the old VxWorks operating system into IOS. Thus, in this subsection, the focus is on VxWorks for the 350 series only. This is because when we talk about the 1200 and 1100 series, it is obvious that they only use IOS (IOS is discussed later). Connecting to the access point can be accomplished in a number of ways. First, to access it without any knowledge of its setup, one needs to access it though the console. This is the RS-232 connection on the back of the device shown in Figure 10.10. A serial DB-9 cable comes with the access point. It is blue and should have two ends that look alike. To note is that most Cisco equipment comes with what is called a rollover cable. This cable is used to access the equipment through the console port. The 350 series access points are one of the few Cisco devices that have a different cable for console management than the normal Cisco console cable. Once this cable is connected to both the access point and an open Com port on a workstation, a terminal emulator program must be open. One emulator that is part of Microsoft Windows is called hyper terminal. Other emulators can be downloaded from the internet or purchased. Make sure the terminal emulator program is correctly configured following the details below. To connect a workstation correctly to an access point, the setup on a terminal emulator program needs to have the following settings: Com Port: set this to 1 or the com port to which the cable is connected Bits per second (baud rate): 9600 Data bits: 8 bits Parity: no parity Stop bits: 1 bit Flow control: Xon/Xoff



158


Figure 10.12 Cisco 350 VxWorks console screen.

Once the cable is connected and the terminal emulator program is set up correctly, try to launch the session and connect the access point. At this point, one should see a cryptic-looking menu system similar to that in Figure 10.12. This is the VxWorks operating system. One can navigate by typing the first characters of a command into the window. The commands have brackets around them; some of them have different navigation keys than their names imply. Look out for what is typed in the brackets on the screen to make sure that the correct information is entered.



Wireless Access Points 159

Having learned how to connect to the Cisco 350 access point via the console, there some other ways of setting up and managing this device. This device has a built-in Web interface like the one on the Linksys access points. To access it, one must know the IP address of the unit so one can enter that into a Web browser. By default, all Cisco access points get their address via DHCP. This makes finding the access point’s IP address difficult the first time. Cisco stepped in and created a tool called IP setup utility (IPSU) that can find the access point’s IP address from its MAC address. For this tool to work, it must be installed on a workstation that is on the same network segment as the access point. Once this software is installed, one can launch the application, type in the MAC address of the access point, and it will show the IP address. The MAC address of any Cisco access point is written on the back of the unit. Now that the IP address is known, connecting to the access point is possible through the Web. To do so, just type in the http:// and the IP address of the access point. This will bring up the Web interface. For VxWorks, this method is the preferred method of configuration. One key advantage that Cisco has made for itself in the large enterprise space is the ability to create configuration scripts for almost all of its products. With VxWorks, still being an Aironet/Cisco code, this was not achieved. This meant that the ability to easily script the configuration in VxWorks was rather difficult. Looking at Figure 10.13, one can see what the Web interface looks like for the VxWorks operating system. The final method of accessing the access point’s management functions is through telnet. This has a very similar look and feel to the console although it can be done remotely from any connection with IP connectivity. In some newer versions of code, Secure Shell (SSH) can also be used. Telnet is prone to eavesdropping because its authentication takes place in cleartext. To telnet into an access point, all one needs is a telnet program. UNIX and Windows both have telnet ability right from a command line or shell. Most Cisco equipment can perform a telnet action from one device to another.

10.2.2 Cisco 1200 Series Access Point Now to the Cisco 1200 series access points. These all come in a plenumrated metal case; the case is shown in Figure 10.14. These access points have the latest and greatest features available from Cisco. They are capable of supporting 802.11b, 802.11g, and 802.11a simultaneously. To support 802.11a, a paddle card must be installed into the access point. This card is shown in Figure 10.15. Delving deeper into the access point itself, one can see from Figure 10.14 that the access point has the same LED layout as the 350. Each



160


Figure 10.13 Cisco 350 VxWorks Web access screen.

Ethernet activity, association status, and radioactivity light performs the same function as detailed above for the 350 series. Looking at the connections on the device in Figure 10.16, one can see that the 1200 access point has two antenna leads, a power connection, an RJ-45 console port, and an RJ-45 Ethernet access port. One big difference between the 350 and 1200 series is the more common use of the correct Cisco-style console cable. The 1200 access point has the same Cisco console cable that almost all other Cisco equipment has. Another big difference is the directly connected power port. This is now included in this model, unlike the 350, which required POE from a switch or injector. The 1200 series access point is capable of POE and can be plugged in with direct power from a power cord. This allows for maximum flexibility with regard to powering the device. Also, all currently shipping 1200 units from Cisco are 802.3af compliant. The 1200 series access point was made with the capability to upgrade to new technologies. This was often a requirement of customers. They wanted the capability to support technologies that were close to being released, although not available at the current time. To make sure that




161

Figure 10.14 Cisco 1200 access point detail.

they did not have to replace all their wireless networks, most customers required that the access point have easy upgradeability, to include support for other, newer technology. In the case of the Cisco 1200 series access point, this change was as easy as swapping out one card for another. This allowed the device to go from 802.11b to 802.11g rather easily. When the 1200 first came out, 802.11g was still being standardized. This meant that Cisco had to release the 1200 with 802.11b technology. To upgrade to 802.11g, a new radio must be installed. Cisco wanted to make the transition from 802.11b to 802.11g an easy one and did so by making it easy to change out the radio cards. Looking at Figure 10.17, one can see how the radio comes out. The 1200 series access point runs IOS. Most Cisco products run this operating system. It is laid out the same way as other Cisco devices. This means that anyone familiar with a Cisco router or IOS-based switch should be able to pick up the access point IOS without too many issues. This IOS still has the ability to allow connections from the console, telnet, Web browser, and SSH. One of the big differences in IOS versus VxWorks is the ability to script configurations. With IOS, text configurations are easily



162


Figure 10.15 Cisco 1200 removable 802.11a radio detail.

Figure 10.16 Cisco 1200 side port detail.

put into and pull out of any 1200 IOS-based access point. This means that basic standards can be inserted into engineering templates and some assurance can be given that all access points will share these common settings. Figure 10.18 reveals the IOS Web-based interface. It looks very different from the VxWorks. It has all the same functions of VxWorks, although it is




Figure 10.17 Cisco 1200 radio replacement.

Figure 10.18 Cisco 1200 IOS-based web access screen.


163


164


laid out a little neater. When using this configuration tool, one of the quickest ways to get the access point up and running is the express setup page. When one consoles or telnets into an access point, one gets what is called a command line interface (CLI). This is very similar to a Cisco router. Once logged in, one is in user mode, which is a very restrictive mode and allows for only simple viewing commands. To get into the next privilege mode, called the EXEC mode, one must type “enable” into the command line. This should change the prompt from > to #. Now this mode can run all viewing commands. If one wants to make a change, one must enter another mode, called global configuration mode. This mode allows one to make changes. To get into global configuration mode, one types “configuration terminal” into the command prompt. One can make changes that apply to the entire device, including changing the device’s name, setting the clock, adding users, etc. The final menu type is called interface configuration mode. This menu allows one to perform actions on certain types of interfaces, such as the 802.11b radio, or a particular VLAN. To get to the interface configuration menu, one must already be in global configuration mode and then type “configure interface dot11radio0.” This gets you into the configuration menu for the 802.11b radio. Whatever designator is assigned to the interface one wants to configure is the one that must be referenced in the command. Figure 10.19 provides all the details and shows what the IOS command line looks like. Looking at this figure, one can see that one of Cisco’s features is support for the tab command. In most UNIX operating systems and Cisco IOS, the tab key can be used to complete a command. As one can see from Figure 10.19, to enter enable mode, all that is needed is “en.” This can also apply to other menu items such as global configuration mode; typing in “config t” can easily access this mode.

10.2.3 Cisco 1100 Series Access Point The final wireless access point from Cisco discussed herein is the 1100 series. This access point was created for a low-cost, small office access point. It is only available in 802.11g or 802.11a. Unlike the 1200 series, it cannot run two radios at once. Older 1100 access points ran 802.11b standard. To upgrade from 802.11b to 802.11g, one only needs a small radio chip. This makes the 1100 a modular upgradeable radio. Both the 1200 and 1100 series access points only need a single screw to get to each of their radio cards. The 1100 series access points do not support external antennas, unlike the 1200 series, which does support external antennas. The 1100 has a 2.2-dBi omni antenna permanently affixed to the unit. This makes the unit difficult to use in places where external




165

Figure 10.19 Cisco 1200 IOS console screen.

antennas would be needed. If one needs an access point that requires external antennas, go with the 1200 series. If the solution is for a small office, then the 1100 series is a more cost-effective option. The 1100 series access point is capable of being powered by POE. Looking at Figure 10.20, one can see that the 1100 series access point looks similar to the 1200, although it is slightly smaller and has affixed antennas. Other than the many physical differences between the 1100 and 1200 series access points, from software perspective the two are very similar. Both run IOS and operate under the general command structure. This makes an 1100 and a 1200 look almost the same from a remote console perspective. This also helps with large companies that manage both 1200 and 1100 series access points. They both look and feel the same from a software configuration standpoint.



166


Figure 10.20 Cisco 1100 access point detail.

10.3 Chapter 10 Review Questions 1. What is the correct DC voltage used to power a Cisco 1200 series access point when using POE? a. −23 b. –34 c. –99 d. −48 2. The cable used to connect to a Cisco 1200 series access point is commonly referred to as a ________________. a. Patch cable b. Coaxial c. Roll down d. Roll over




167

3. What administrative methods are supported in a Cisco access point? a. IPSSU b. Console c. Telnet d. Web based 4. Which of the following is not a connection on the Cisco 1200 series access point? a. RJ-45 b. Power c. Fiber d. Ethernet 5. Which types of antennas does the Cisco 1100 series access point support? a. 13.5 dBi Yagi b. 5.5 dBi patch c. 2.2 dBi omni d. 6 dBi patch 6. What administrative methods are supported in a Linksys access point without firmware modifications? a. IPSU b. Console c. Telnet d. Web based 7. What command should be issued on a Cisco IOS access point after accessing it from a telnet or console to reach the enable mode? a. Int Dot11radio b. Config t c. Enable d. Configuration mode 8. What types of operating systems do new Cisco access points use? a. IOS b. VxWorks c. Microsoft d. Linux



168


9. Do Linksys access points support POE? a. Yes b. No 10. Does Cisco currently sell 350 series access points? a. Yes b. No 11. Cisco 1200 series access points are capable of supporting 802.11b, 802.11g, and 802.11a all at the same time. a. True b. False 12. What standard do all Cisco 350 series access points adhere to? a. 802.11a b. 802.11b c. 802.11g d. 802.11n 13. What Cisco access point runs the VxWorks operating system? a. 1200 b. 1100 c. 350 d. 1400



Chapter 11

Wireless End Devices When looking at wireless end devices, one notes that laptops are far from being the most common. Depending on how wide the defi nition of wireless end devices is, cell phones could easily top the list. Back in the pre-802.11 days, most wireless devices were handheld scanners used for inventory tracking. As technology advanced and data rates increased, a new market for wireless laptop communication evolved, thus adding wireless laptops to the list. With the continued advancements in wireless technology, more and more devices began to incorporate wireless connectivity. During this time, PDA devices began to hit the market with wireless capabilities. Today we have multiple communication capable devices. These smart phones and PDA devices have Wi-Fi, cellular, Bluetooth, and IR technologies that can all be used to transmit data to and from a single device. This chapter looks at the most common wireless end devices, explaining how they work, what purpose they serve, and how security relates to them. After exploring all the wireless devices, this chapter then explores what threats face each of these devices and how one can protect them from those threats. Before getting into the specifics of each type of wireless device and their related risks, it is necessary to explore the risks of wireless devices in general. Almost all wireless devices are mobile; unfortunately, this means there is a risk of lost or stolen devices. Mobile devices such as laptops, PDA devices, and others are often misplaced or, worse, stolen. This can mean the device itself is stolen or the data inside it is stolen. When this happens, the loss of an asset has occurred. Depending on the 169 © 2006 by Taylor & Francis Group, LLC


170


price of the device, and the data that may have been on it, losing these devices can be a real risk to any organization. To protect against this risk, companies can use encryption software for the protection of their data and good policy development to combat device loss. The most important risk-reducing action is education and awareness.

11.1 Laptops Laptops are the devices people most commonly picture when they think of wireless LAN end devices. Laptops are able to perform all the same functions as a desktop computer. Many people prefer laptops to desktop computers because they are small, portable, and convenient. Once the wireless industry took off, the laptop industry started to see the benefits of going wireless. More and more consumers also began to see the benefits; they could browse the Internet from their couches or anywhere in their homes without the inconvenience of wires. Today, the majority of laptops come with a built-in wireless adapter. Laptop computers connected to wireless networks face not only the security threats all network connected computers face, but also the added threats from wireless access. One example of a threat that wireless laptops face is a hotspot. Using a laptop on a hotspot might seem safe, but the truth is, it is far from safe. The way most computers communicate on a network is by network protocols; some of these protocols were made back in the early days of the Internet itself. The age of these protocols can lead to a host of security issues. This is magnified when using a hotspot, especially one that does not use encryption. These older protocols are more exploitable in a wireless environment than in the physical media of a wire. Picture the following example of how wireless laptops are at risk. A person goes to his local hotspot. It could be in an airport, coffee shop, bookstore, or practically anywhere. The laptop user fires up his laptop and connects to the network. The user fills out the security information necessary to gain access and the hotspot proxies all his requests to the Internet. Because the hotspot is doing the security at the access point, the airwaves are not secure in any fashion. If a hotspot vendor was to secure the airwaves with some form of encryption, it would be more of a risk to the vendor than an advantage for its customers. If the hotspot vendor says his network is secure and an incident happens, the vendor could be at fault and liable for all damages. Because of this liability, most hotspot vendors protect only Internet access and leave the airwaves open. Now that the laptop user has connected to the Internet, he will most likely want to retrieve his e-mail. The most common e-mail retrieval protocol is POP3 so one assumes that is what he uses. The POP3 protocol



Wireless End Devices 171

has quite a few security deficiencies; for example, it sends the username and password in cleartext. This means that anyone in the area using a network sniffer can capture the username and password of someone else downloading mail. Once the name and password have been captured, the attacker is able to log into the laptop user’s mail system and read, write, or delete his e-mail. Other protocols have problems similar to POP3. This book explores a number of weak protocols.

11.2 Tablets Tablets have slowly begun pushing their way into the marketplace. Many industries were unable to function with the small form factor of PDAs but did not like the weight or two-handed use of laptops. Tablets were created to mimic a notebook with the advantage of having all documentation and handwriting electronic. For example, this has helped healthcare become more streamlined. Tablets have moved forms and documents that need signatures from paper base to electronic factor. The tablet is nothing more than a computer with a compact form and an electronic pen. This means it is also susceptible to all the threats mentioned in the above sections. An example of a threat related directly to tablets involves the use of electronic signatures. When a tablet is used to electronically sign a document, the signature becomes software based. This means the signature can be cut, copied, and pasted into a different form. This can mean that someone can easily forge signatures. This threat is not directly related to wireless use; however, it is very common on tablet computers.

11.3 PDA Devices Information management tops the list of features for PDA devices. For the majority of the 1990s, PDA devices were marketed as only information managers. As the turn of century approached, these devices began to incorporate new technology that pushed their features way beyond that of a personal information manager. With the advent of wireless technologies, PDA devices have become useful for multiple business applications. Some of these involve database querying, record keeping, record viewing, and other real-time tools. These new abilities make PDA devices practical and cost effective for many enterprises. All PDA devices run on some type of hardware; this hardware enables the device to turn on and use various wireless technologies. Today, there are so many PDA device manufacturers that talking about each of them would easily fill this book. Having all these manufacturers creates both



172


positive and negative effects. On the positive side, having a large selection is always good for consumers. This helps customers find better prices and allows for a wide variety of features. Allowing the consumer to choose the type, speed, form factor, and available transmission technology allows for a quicker purchase. However, supporting different devices can cost a lot more money than supporting a single one. This has become evident in corporate environments where personal PDA devices are accepted. This culture has brought with it requests from users to have IT personnel to support these devices. Unfortunately, when a company tries to save money by not embracing a corporate PDA roll-out and instead provide end users with the ability to use their own personal PDA devices, the support cost alone will outweigh the hardware costs of an enterprisewide roll-out. With multiple vendors running multiple versions of software, being able to support this gamut of devices is overwhelming. Even with all the different hardware options out there today, across all the different vendors, none of these devices would work without software. This software allows the device to power on and allows it to interact with the user. Similar to the hardware market, the software market offers a wide variety of options.

11.3.1 Palm The company that makes Palm software was created in 1992. One of their first devices was called a Zoomer. The Palm Company has changed hands several times since its creation in 1992. It was acquired by U.S. Robotics in 1995, then by 3Com in 1997 as they acquired Palm’s newly appointed parent company U.S. Robotics. Then, three years later in 2000, 3Com turned Palm back into an independent company, bringing it back full circle to where it began in 1992. In 2002, Palm spilt its operations into two companies owned by the parent company Palm: (1) Palm Source, which created and licensed the Palm OS software, and (2) Palm Solutions Group, which handled the creation of hardware for the Palm operating system. Today, Palm represents a large portion of the handheld market. However, its market share began falling due to the recent advancements of other companies. Palm devices continue to be primarily personal information management (PIM) devices carrying the user’s contacts, calendar, and other useful tools. The base Palm software comes with the following applications: Address book Mail



Wireless End Devices

173

Date book To-do list Hotsync software Some other applications available with newer models include: Note Pad Short Message Service (SMS)

11.3.2 Microsoft CE and Pocket PC Another software maker for handheld devices is Microsoft. Microsoft made a name for itself with the creation of the computer software operating system called Windows. In 1996, Microsoft started introducing its successful desktop software into the handheld computing market. The first version of software released by Microsoft was CE 1.0. This software was similar to Windows 95 and had the same look and feel. CE 1.0 was a slimmeddown version of Windows 95 that was made solely for handheld devices. As time went on, newer versions of CE began to emerge from Microsoft and new initiatives emerged as well. One of those initiatives was an OS targeted at wireless handheld computers and mobile phones. This initiative created Pocket PC. This left Microsoft with CE and a choice: either keep it going or replace it with Pocket PC. Microsoft took its version track for CE and added it into part of their embedded operating system group. This meant that for devices like bar code scanners, radios, TVs, and other similar devices, the Microsoft CE platform will be an available standardized solution. This split between CE and Pocket PC allowed the Pocket PC group to build more functionality into its smart phone operating systems and PDA devices. Pocket PC was geared more toward a user who needed to perform multiple functions like e-mail, contact management, calendaring, and many other PIM type applications. On the CE platform, the normal user would only be using the device for a particular function, such as bar code scanning or package delivery. Microsoft allows hardware makers to customize their CE software to operate on their devices. This has helped companies create devices and not software. Many devices have hardware built by a certain company and the software provided by Microsoft. One example is a portable handheld scanner that runs CE. A hardware company makes the scanner and Microsoft provides the software. Looking at CE today, which is now CE.NET, when compared to Pocket PC, the main difference lies in the PIM software. CE.NET has no PIM software included; there is no calendar, e-mail client, or task manager.



174


This is because of the change Microsoft made from handheld devices to devices in general. The Pocket PC version that Microsoft released is made just for PDAs and smart phones. Microsoft has talked about splitting smart phone software away from Pocket PC software. The real issue that has stopped this event is the massive push to condense these functions into a single information management and communication device. Currently, if one purchases a PDA unit, it should come with Pocket PC and not Windows CE; if it does not, be careful as this means that the manufacturer is way behind the curve. Some of the applications that come standard on the Pocket PC are listed below. Make a note that the smart phone OS has only a limited subsection of the applications listed below. Standard applications that ship with Pocket PC include: Pocket Outlook: – Contacts – Calendar – Tasks – Inbox – Notes Microsoft Pocket Word Microsoft Pocket Excel Microsoft Reader MSN Messenger Terminal Services

11.3.3 BlackBerry RIM OS BlackBerry’s RIM operating system encompasses software, hardware, and the needed application to connect the mobile device to corporate e-mail servers sold, all from a single company. The device OS was created from the ground up to have an always-on connection to the Internet and email. Because of this, a solution had to be created to support this type of connection to a corporate e-mail server. Most companies had pieces and parts to this whole solution. BlackBerry created an entire solution based on its own products, which was never done by a single company before then. This solution is based on all of their own hardware and software. There are some major differences between the BlackBerry devices and other PDA versions. First, they have no stylus to navigate; they use a roller on the side of the device to facilitate navigation. They also all come with a small keyboard that allows for easy messaging. Some PDA hardware makers have seen these differences and have adapted some of them to their offerings, such as the keyboard for Pocket PC devices. © 2006 by Taylor & Francis Group, LLC



175

11.3.4 Symbian OS The Symbian OS was created to meet the needs of users who were complaining about how the industry failed to respond to the need for smart phones that can combine PDA, messaging, and voice calling functions on a single device. Most manufacturers or software makers took whatever product they already had and adapted it to fulfill such needs. This meant PDA devices and software tried to tie messaging and voice capabilities into one device. On the other side, phone makers tried to create more PIM-type applications on their devices and in their software. Then a couple of these companies got together and led a joint effort to create software that was originally created to operate on smart messageand voice-aware devices. This led to the creation of the Symbian OS.

11.3.5 Linux Linux, which provides free-of-charge, community-based operating systems, has only found its way to a few devices. Many advanced users have found ways to port Linux onto their devices, although few manufacturers support or sell a Linux PDA. These devices provide an alternative to licensing software from other companies. The Linux PDA has still a long road ahead before one sees them out in any significant numbers on handheld devices.

11.4 Handheld Scanners In this section we are going to talk about handheld scanners and barcode readers. These devices are from pre-802.11 days when the primary use of wireless was to perform inventory tracking or barcode reading. Some of these devices run on 900 MHz and are way outside 802.11. If you have ever been to a large retail store, you have probably seen the handheld scanner guns the cashiers use to scan items. Most modern-day scanners use wireless networks to operate and function in the rule set of 802.11b. The handheld scanner is also used widely in warehouses or distribution centers. One big issue holding back these kinds of devices is battery life; several vendors have decided to wait or to not pursue 802.11g or 802.11a because of the additional battery capacity needed to maintain their current runtimes. This means that many of these devices will not move to 802.11g or 802.11a. Another major issue with these devices is the chip set and the ability to handle a more processor-intense security method such as 802.11i. Performing Advance Encryption Standard (AES) on these devices will either not be possible or will dramatically shorten the battery life. © 2006 by Taylor & Francis Group, LLC


176


The threats to these devices specifically deal with the lack of advanced security capabilities. Some of these products are stuck with only supporting WEP, which can mean a major risk in itself. Another interesting note in this regard deals with 802.11 FH systems. Some have stated that these systems have a lower risk associated with them, stemming from the fact that one cannot go out to a local computer or electronics store and buy an FH card. This means fewer people are likely to attempt to circumvent any secure mechanisms already in place. This topic must be addressed when weighing risks versus controls of older systems such as the ones that support old technologies.

11.5 Smart Phones Today, smart phones are becoming everyday devices that can serve as a planner, phone, Web browser, and small computer — an all-in-one device. Depending on what the manufacturers call the device, it can be a smart phone or a PDA phone. The industry has classified these devices based on their main function, with smart phones being a phone first and PDA second, and PDA phones being a PDA first and a phone second. This section addresses these devices as smart phones although the industry has made a clear delineation between hardware and software types. Microsoft, for example, has two main flavors of operating systems: Smart Phone OS and Pocket PC OS. Both operating systems perform the same functions, with the only differences being that the Smart Phone OS was developed specifically for a phone and the Pocket PC OS was originally developed for handheld PDA devices and later adapted by each hardware manufacturer to incorporate phone functions. These devices bring a slew of complexity and risk into account when using them. The first threat directly related to smart phones is the multiple connectivity options. One of these devices can use Wi-Fi, Bluetooth, and a cellular technology on the same device. When security threats come out and are directed at one of these technologies, the user must protect against each of these threats independently. This means that the user must have a higher level of knowledge and understanding about how each of these connection technologies works and is secured. The virus threat has plagued computers since November 3, 1983, when Fred Cohen conceived the first computer virus. Today, over 20 years later, there are more than 50,000 viruses crawling the Internet. Just recently, we have started to see viruses directed at mobile devices such as smart phones or PDA phones. The virus threat was always known on these devices, although it was only related to having a mobile device carry a




177

computer virus and then, in the process of synchronizing the mobile device to the computer, it would infect the computer with the virus. With viruses and smart phones, not only are there the issues of removing malicious code, but there is also the threat of having malicious code place calls. A virus can attack a device, infect it, and make that device place phone calls or make requests to the Internet that can cost the user money. Some of these devices have means built directly into them to combat viruses. The Microsoft Smartphone and Pocket PC OS have the default ability to prompt the user before running any executable code. While this may protect against viruses that are sent directly to the phone, most mobile virus writers are well aware of this and have changed their method of attack. The real threat has now moved from viruses more toward Trojan horses. The mobile virus writers want to get people to accept the executable prompt, inserting the virus into a game or an application the user might want to use does this. When they press OK to the prompt, they allow the Trojan to execute and deliver its malicious payload. Another amazing feature of some smart phones is GPS. GPS is used for a couple of things on a cellular phone or smart phone. First, a method of locating a cellular phone is needed for emergency services. This is one requirement that has put GPS on most cellular phones. GPS on cellular phones most likely is not true GPS, because true GPS does not work correctly indoors due to a line-of-sight requirement between GPS satellites and the GPS receiver. Some cellular carriers use a technology called gpsOne, which works by having the phone be capable of receiving both a GPS constellation signal and a wireless network signal. The wireless network signal is also broadcasting GPS information to the device. Once the device picks up these signals, the measurements are combined by the location server to produce an accurate position fix. Either way GPS is used on a phone, the risks remain the same. The first risk that comes to mind is tracking. Some information on GPS tracking was defined in the GPS section of the wide area technology chapter (Chapter 7, Section 7.2). In that chapter, one learned that the GPS system itself could not be tracked. The only way to track a GPS receiver is for another technology such as cellular to transmit the location information. Most phones have not only GPS for emergency services, but also location-based services (LBS). Location-based services offer applications the ability to use real-time location data; this can be used to track a delivery driver or find the closest taxi driver to someone who just called in. As amazing as LBS sounds, the malicious implications as well as privacy concerns can be major issues. Luckily, some phones have the ability to disable this feature. There have been a number of what-if scenarios discussed about the future of LBS on cell phones. The first one is as follows. An attacker could



178


shut down cellular coverage in a given area by sending a large number of SMS messages to people who are in a given area. This is done using LBS on top of the SMS threat. Another what-if is the LBS virus threat. What if an attacker can disable, steal, or erase someone’s phone when that person is in a particular place? Another more real-life threat relating to smart phones is the fact that most of these devices have the ability to support a direct connection to the Internet. This means that these devices are out on the hacker-plagued Internet. The real threat comes from a hacker hacking into the device.

11.6 Wi-Fi Phones Wi-Fi phones use the network for phone calls and require optimal bandwidth to operate. They are used to locate employees who work in a large area or who do not have a desk. They can also be used to roam from a Voice-over-Internet-Protocol (VoIP) network to a cellular network. Wi-Fi phones have a major problem with regard to security in the fact that security processing takes time. The time used for security processing causes phones to have bad reception or drop calls. This has created another risk with Wi-Fi phones: either buy into a proprietary solution to increase the security processing of your network or lower the security to accommodate the Wi-Fi phones.

11.7 Chapter 11 Review Questions 1. What program is NOT in CE.NET but is in Pocket PC? a. Activesync b. Windows Media Player c. Outlook d. Internet Explorer 2. What was one of the first PIM devices? a. PLAM b. CE.NET c. Smart phone d. Zoomer 3. What is one of the biggest risks facing wireless devices? a. Theft b. Hacking c. Information loss d. Data corruption © 2006 by Taylor & Francis Group, LLC



179

4. Healthcare has seen a great increase in productivity due to which wireless end device? a. Laptop b. Tablet c. Smart phone d. Wi-Fi phone 5. Windows’ first version of CE was based on __________. a. Windows NT4 b. Windows 2000 c. Windows 95 d. Windows 98 6. What technology do most smart phones have that can be used to track them? a. CDMA b. CE.NET c. GPS d. PocketPC 7. When was the first computer virus written? a. 1980 b. 1990 c. 1983 d. 1999 8. What security feature has Microsoft put into its Smartphone OS and PocketPC OS? a. Security filtering b. User action required to run an executable file c. Anti-virus d. Firewall 9. What makes securing a wireless end device more difficult? a. The large number of device types b. Multiple connection options (e.g., WLAN, PAN, cellular) c. Loose security d. Extra batteries 10. What does the acronym “LBS” stands for? a. Lucky basic security b. Location-based security c. Location-based services d. Location basic service © 2006 by Taylor & Francis Group, LLC


180


11. Which operating system was created from the ground up as a PDA, phone, and PIM combination? a. Smart phone b. RIM c. Symbian d. CE.NET 12. What is a major reason why handheld scanners have not gone 802.11g? a. Security b. Speed c. Battery life d. Chip set availability



Chapter 12

Wireless LAN Security Wireless communications were always prone to security issues well before any of the 802.11 standards. Most people never thought about wireless security until the market responded with news, ads, and products to make the public aware of the dangers. What makes wireless such a security threat has to do with the fact that it is wireless. This means that data is transmitted over airspace and is susceptible to eavesdropping by anyone in a given area. Over the years, different types of encryption have been used to protect the data inside this transmission; however, this has not always been successful. This chapter walks through the history of wireless local area networking security, seeing how the industry went from welcoming wireless with open arms to fighting it day by day as people took it upon themselves to deploy it. This chapter looks at the history and then identifies most of the security methods available today. After reading this section, one will have a clear understanding of the methods in use in the past and present to protect wireless networks.

12.1 Wireless LAN Security History Looking at the history of wireless networking, there are a number of interesting points, articles, attacks, and tools that were released over time. This section reveals how many of these wireless security methods were created, broken, and improved. This helps us understand how the market reacted to each new security solution and its subsequent exploitation.



182


On October 27, 2000, Jesse R. Walker published one of the first articles detailing the risks and issues regarding security in wireless 802.11b networking. In this article, he made comments about wireless networks being similar to data drops in the parking lot for anyone to use. He also pointed out a key point about increasing the key size of the WEP cipher during a time when the industry was pushing to increase the size of WEP keys from 40 bits to 128 bits. Walker pointed out that increasing the size of a WEP would not help against many of the attacks that were known at that time. An article entitled “Your 802.11 Wireless Network Has No Clothes,” published by the University of Maryland, was released on March 30, 2001. It contained details about a flaw in shared key authentication. This flaw took advantage of a technique widely known as plaintext cryptanalysis. This exploit worked by having access to a cleartext piece of data and its encrypted version. When comparing these the two, one can derive the key used to encrypt it. On a wireless network at that time, this key was used for most communications across these networks. On July 16, 2003, Scot Fluhrer, Itsik Mantin, and Adi Shamir presented a paper titled “Weaknesses in the Key Scheduling Algorithm of RC4” at a security conference. This paper is one of the first papers to exploit the WEP encryption method itself. It stated mathematically how to break a WEP key, although it also states that the authors never tried it in a practical lab test. This paper was the main reasoning behind the wireless industry considering WEP unacceptable and in need of replacement. After the release of “Weaknesses in the Key Scheduling Algorithm of RC4”, members of AT&T decided to perform a real test on WEP encryption itself. This was because the author of the paper detailed how to do it without stating how they really did it. So the AT&T Lab members created a report, Labs Technical Report TD-4ZCPZZ, “Using the Fluhrer, Mantin, and Shamir Attack to Break WEP.” This test was successful in breaking WEP encryption. Interestingly enough, the day this report was published the first WEP cracking tool, called WEPCrack, was also released. This tool was a Perl-based script tool with the ability to crack WEP keys. Five days later on August 17, 2001, Airsnort, a GUI WEP cracking tool, was released. For some reason, many members of the press acknowledged Airsnort as the first WEP cracking tool, when in actuality it was not. On August 21, 2001, the “Weaknesses in the Key Scheduling Algorithm of RC4” paper was publicly released. After this paper was released, the AT&T folks had an issue with their lab results. It would be difficult to have a lab test results published before publishing the theory they were testing. This led to an update; in order to follow up with the “Weaknesses in the Key Scheduling Algorithm of RC4” public release, the AT&T testing paper was updated to reflect the correct date.



Wireless LAN Security

183

On February 6, 2002, The University of Maryland published an article entitled “An Initial Security Analysis of the IEEE 802.1x Standard.” By this time, the industry’s answer to the WEP security concerns was to move to an 802.1x security method. This article detailed many issues surrounding the 802.1x standard and the lack of two-way authentication. This gave hackers the ability to perform a man-in-the-middle attack on 802.1x wireless networks. After the WEP flaw was released and tools were now considered mainstream, we started to see a number of networks being comprised, one of which involved a member of the press and a network security professional. They were driving around demonstrating a new trend called war driving. During this war drive, they went up and down a number of streets in Silicon Valley. Some of the networks they found included Sun Microsystems and Nortel Networks, both of which did not even have WEP enabled. Later, when the story hit the press, Sun Microsystems commented that the network was a test lab that was not connected to any part of its internal infrastructure. Nortel refused to comment on the press release. On May 1, 2002, Best Buy had an encounter with wireless hackers. Best Buy was using wireless POS systems, which increased its ability to move its register’s layout during peak purchasing times, such as Christmas. Some store POS systems were set up with the wireless services encryption enabled. This meant that when a customer used his Best Buy credit card, the number went across the airwaves to the corporate server for validation. Some wireless hacker sitting out in the parking lot was able to capture this traffic and get the numbers of multiple people’s credit cards. On September 22, 2002, Cisco released a response about the weakness of 802.1x cited by the University of Maryland; this article, entitled “Cisco Aironet Response to University of Maryland’s Paper,” stated that if one uses Cisco’s proprietary method of authentication called LEAP, one would not have the problems pointed out in the previous University of Maryland article. This was because Cisco had two-way authentication between both the client and the access point. This meant not only was the client authenticating to the network, but the network was also authenticating to the client. On November 13, 2003, Paul Timminsand and Adam Botby were arrested by the FBI outside a Lowe’s home improvement store in Michigan. They were accused of “breaking into” a number of Lowe’s stores across the Midwest. They were breaking in through the wireless networks and from there they were trying to hack into a mainframe computer located at Lowe’s headquarters. On November 27, 2003, a tool came about that could exploit LEAP. This tool, called anwrap, allowed a user to try multiple authentication attempts to the access point. This was easily combated using a password



184


lockout policy, forcing an account to lock after a predefined number of unsuccessful attempts. This led to one of the latest tools called asleap. This tool was released on April 8, 2003, by Joshua Wright. It is also a LEAP attacking tool, but unlike the one previously released, this one has the ability to perform offline dictionary attacks on LEAP passwords. This tool prompted Cisco to abandon LEAP in favor of a new protocol called EAP-FAST.

12.2 Authentication When connecting to a wireless network, one must perform some type of authentication. There are two main types of authentication per the current IEEE standards: share key authentication and open key authentication. The next section will detail both authentication types, detailing how they operate and what security is present in their use.

12.2.1 Shared Key Authentication Shared key authentication was created to be the more secure of the two types; however, as we will shortly see this actually became the less secure due to a small oversight in how it validates user keys. To understand shared key authentication the section below will explain how it works. Shared key authentication works via a challenge response mechanism. To explore this process, one must first connect to the network. Having the client device send out a probe frame does this. This frame will look for available wireless networks and their connection settings. Once an access point “hears” a probe, it will respond with a probe response frame. This frame will identify all of its connection settings to the end device. In some cases, an end device will hear many responses from different access points in the area. To make sure that the end device connects only to the access point with the best signal, the probe response frame has a value for current signal strength. A client might hear multiple replies, although it will only connect to the access point with the highest signal strength value. Once the end client hears this and deter mines that it supports the same settings as the access point, the next portion (called authentication) takes place. When the end device wants to authenticate, it sends an authentication response frame to the access point. This frame is evaluated; once the access point determines it is an authentication request, it will send a challenge packet back to the client. The challenge packet consists of a cleartext piece of data. The end device is required to encrypt this data with its WEP key and sends it back to the access point. Once this is done and the access point receives the packet, it checks it against what it has for the encrypted version of that packet. If the results match, the access © 2006 by Taylor & Francis Group, LLC


Wireless LAN Security 185

Shared Authentication

Laptop Client sends probe. Access points respond with a probe response. Client evaluates probe response and sends authentication request packet to best access point. Access point send a unencrypted challenge packet. Client uses it’s WEP key to encrypt challenge packet and sends it back to access point. Access point sends authentication response packet.

Figure 12.1 Shared key authentication.

point will allow the end device onto the network. If the results do not match, the authentication fails and the end device is denied a connection. To get a full understanding of how this works, look at Figure 12.1, which outlines the connection and authentication process.

12.2.2 Open Key Authentication Open key authentication was originally seen as less secure than shared key authentication. The intent was to make an open network, thus not requiring clients to have knowledge of the WEP key. As security became an increasingly visible issue, many vendors turned to the drawing board. This created a real problem: coming up with a solution that improves security while staying within the standard guidelines. These efforts led to the idea of using open authentication and, unlike before, this open authentication would require the use of a WEP key. When used, the WEP key was required to connect to the network. This worked because when one talked with the right WEP key, one’s cyclic redundancy check (CRC) passed its test and the frame was allowed to access the network to it destination. Looking at how open authentication works, one sees that the end device connected to the network as it did with shared key. It makes a probe request, listens to probe responses from multiple access points in the area, and then determines the best access point to make a connection with based on signal strength. © 2006 by Taylor & Francis Group, LLC


186


Open Authentication

Laptop Client sends probe. Access points respond with a probe response. Client evaluates probe response and sends authentication request packet to best access point. Access point sends authentication response packet.

Figure 12.2 Open key authentication.

How do open and shared key authentication differ? Open authentication sends an authentication request but does not receive a challenge; instead, it is allowed to talk by default. When WEP is enabled, the process is slightly different. When the wireless client starts to talk, it automatically encrypts all the data with WEP encryption. When the access point hears data being sent, it decrypts the frames and forwards them. If the frames are encrypted with a different key than the access point, the decryption portion fails and the packet is dropped. To get a better understanding of how open authentication works, as well as how it differs from shared key authentication, look at Figure 12.2.

12.3 SSID When one looks at security with respect to the SSID field, one sees that most networks, by default, broadcast this information to anyone who is listening. As more people started to take a deeper look into the security of wireless, they considered hiding the SSID in beacon frames. It was noticed that if the SSID was not broadcast, the existence of a wireless network could be somewhat masked. This masking would require the client to send a probe for any available wireless networks. In most IEEE wireless networks, the existence of the SSID is easily attainable with some sort of wireless sniffer. This is because the SSID is part of the process of connecting to a wireless network. Looking at Table 12.1, one can see that the SSID is a present in the header of a wireless probe request frame. This information can be read by any sniffing program, thus defeating any




Table 12.1

187

SSID Sniffer Capture

802.11 Beacon

FC=........,SN= 448,FN= 0,BI=100,SSID=,DS=11

802.11 Probe Req

FC=........,SN= 689,FN= 0,SSID=AEE

802.11 Probe Req

FC=........,SN= 690,FN= 0,SSID=AEE

802.11 Beacon

FC=........,SN= 449,FN= 0,BI=100,SSID=,DS=11

802.11 Probe Req

FC=........,SN= 691,FN= 0,SSID=AEE

802.11 Probe Req

FC=........,SN= 692,FN= 0,SSID=AEE

802.11 Probe Req

FC=........,SN= 693,FN= 0,SSID=AEE

802.11 Probe Rsp

FC=........,SN= 451,FN= 0,BI=100,SSID=AEE,DS=11

802.11 Probe Rsp

FC=...R....,SN= 451,FN= 0,BI=100,SSID=AEE,DS=11

802.11 Probe Req

FC=........,SN= 694,FN= 0,SSID=AEE

802.11 Probe Rsp


802.11 Probe Req

FC=........,SN= 695,FN= 0,SSID=AEE

802.11 Probe Rsp


attempts to hide this identification information. Even with the SSID masked, every time a client wants to connect to a network, it will send all of its connections settings, including the SSID, out into the air as part of its probing process. Several vendors have default SSIDs that they program into their equipment. This is one of the first avenues that a hacker will take when trying to exploit a wireless network. Many companies also use very simple SSIDs such as Wireless, WLAN, and BRIDGE. Table 12.2 outlines some of the SSIDs commonly used.

12.4 Wireless Security Basics Before delving into all the different types encryption and advanced security methods, let us take a look at some basic things that can be done to create a secure wireless network. As shown, the SSID does not help security; however, it is still a good idea to prevent its broadcast. In addition, another important thing is to change the default usernames and passwords on all access points. Some even have default WEP keys that should be changed. Next, run open authentication with WEP enabled. Other recommended changes are to prevent DHCP and to segment the wireless network. One key goal would be to develop an information classification policy to educate everyone as to what should and should not cross the airwaves.



188


Table 12.2

Default SSID Detail

Cisco (All Aironet Access Points/Bridges)

– SSID:

tsunami

LINKSYS Product Families:

– SSID:

linksys

Netgear 802.11 DS products, ME102, and MA401Default

– SSID:

Wireless

SMC Access Point Family

– SSID:

WLAN

SMC2682W EZ-Connect Wireless Bridge

– SSID: – SSID:

BRIDGE Wave LAN Family: "WaveLAN Network"

Symbol AP41x1 and LA41x1/LA41x3

– SSID:

101

TELETRONICS WL-Access Points

– SSID:

Any

12.5 Equivalent Privacy Standard (WEP) The Wired Equivalent Privacy (WEP) standard was created to give wireless networks safety and security features similar to that of wired networks. WEP is defined as the optional cryptographic confidentiality mechanism used to provide data confidentiality that is subjectively equivalent to the confidentiality of a wired local area network (LAN) medium that does not employ cryptographic techniques to enhance privacy. This gives us the basic thought of how WEP was created and what goals it originally intended to meet. To meet these goals, wireless had to address the three tenets of information security: (1) confidentiality, (2) availability, and (3) integrity. 1. The fundamental goal of WEP is to prevent eavesdropping, which is confidentiality. 2. The second goal is to allow authorized access to a wireless network, which is availability. 3. The third goal is to prevent the tampering of any wireless communication, which is integrity. To better understand WEP, one needs to look at it more closely. The WEP protocol is used to encrypt data from a wireless client to an access point. This means the data will travel unencrypted inside the wir ed




189

network. The WEP protocol is based on RSA Securities’ RC4 stream cipher. This cipher is applied to the body of each frame and the CRC. There are two levels of WEP commonly available: (1) one based on a 40-bit encryption key and 24-bit initialization vector, which equals 64 bits; and (2) one based on a 104-bit encryption key and 24-bit initialization vector, which equals 128 bits. This protocol has been plagued with issues since its inception. A magnitude of exploits, poor design elements, and general key management problems have made WEP a very insufficient security mechanism. The details regarding the problems of WEP are explored in Chapter 13. This section, however, discusses how WEP operates. One of the original functions of WEP was to have the encryption unable to be affected by loss of the frame due to interference. What this means is when one sends data across the air and loses the frame, there would be no loss to the previous frame. With newer security methods and older wired secured methods, it is common for subsequent packets to have an encryption dependency on the next or previous frame.

12.5.1 WEP Encryption Process As learned in the previous section, WEP uses an RC4 stream cipher to encrypt wireless data. In performing the WEP encryption process, a number of steps are performed. The first step is to generate a seed value. This seed value is used to start the keying process. This value can be referred to as a key schedule or as the seed value. No matter what this value is called, it is considered the WEP key. After this value is defined, it must then be entered into the access point. To ensure that the client can receive and decrypt the transmission, the seed value or WEP key must be entered on to each client. This will allow a WEP encrypted conversation to occur. This value consists of a 26-digit hexadecimal number. This value is not used alone to create a WEP encrypted data stream; a technique to randomize the key is applied as well. This technique uses a 24-bit initialization vector (IV) that is created on a frame-by-frame basis. The technique on which the IV is created differs between vendors. The WEP standard that is outlined inside 802.11b states the IV size and requires that the IV change on a frame-by-frame basis. Outside of this, there are no requirements in the standard defining how to increment or randomize the IV sequence. As one progresses through this book, one will see that the missing definition of an IV sequence is one place where the WEP protocol has a major security issue. Once the IV and WEP key are together, they can be used to encrypt the frame. When the data is r eady for



190


WEP key encryption process Data

IC

Header IV and WEP key

Header

IV

Encrypted

Figure 12.3 WEP process detail.

transmission, the WEP key and the IV are combined; then, using the RC4 cipher, the key and IV are XORed with the data and IVC to create the encrypted frame. Next, a copy of the same IV is put into the frame header as cleartext. The last step is to send the packet. Once the other end receives the frame, the IV is picked out of the frame header and applied to the predefined seed value to produce the same session-based WEP key that was used to encrypt the packet. The same RC4 encryption process is performed in reverse, allowing the encrypted text to turn from ciphertext to plaintext. Once this operation is complete, the CRC is removed and applied to the data to make sure that it was not corrupted in transit. Figure 12.3 shows how the encryption takes place. As one can see, the data and the integrity check are put together linearly as well as the WEP key and the IV. The WEP key and IV are fed to the pseudo-random number generator. The output of this generates a key stream equal to the length of the frame payload plus the CRC. The sender then performs an XOR operation on the key stream and the data, creating an encrypted data stream. The next step involves placing the IV inside the packet header and forwarding the packet to the sender.

12.6 802.1x Both the IEEE and ANSI organizations approved the 802.1x standard. On June 14, 2001, the IEEE approved the standards; four months later on October 25, 2001, the American National Standards Institute (ANSI) approved it as well. The 802.1x standard was designed for port base authentication for all IEEE 802 networks. This means it will work across Ethernet, FDDI, token ring, wireless, and many other 802 networking standards.




191

One thing people tend to become confused about is that 802.1x is in no way any type of encryption or cipher. All the encryption takes place outside the 802.1x standard. For example, on a wireless network, the EAP would use one of its various methods of encryption for authentication. After the user authenticates to the wireless network, they may start a conversation using WEP, TKIP, AES, or one of the many other standard wireless encryption schemes. When looking at the 802.1x standard, at its most basic view one sees actually what it was intended for, port-based authentication. This means the standard takes the authentication request, decides if it is or is not allowed onto the network, and then grants, or revokes, access. Many parts of how 802.1x works are within other standards such as EAP and RADIUS. The 802.1x standard is just a mechanism that denies all traffic except EAP packets from accessing the network. Once the EAP says it is OK for the device to access the network, the 802.1x protocol tells the switch or access point to allow user traffic. This is accomplished by having the network port or, in a wireless situation, each client connection in one of two port states. These states are controlled and uncontrolled. Figure 12.4 reveals the three main designations called out by the 802.1x standard. Each of them has specific rules and functions. The standard was written to incorporate a large amount of different equipment; the names of these functions remain somewhat generic. As one can see from Figure 12.4, the 802.1x protocol leverages two other standards. From the supplicant to the authenticator, the standard is EAP. From the authenticator to the authentication server, the protocol is RADIUS. The 802.1x protocol takes EAP requests, sends them to a RADIUS server, and waits for an answer. Once this answer is received, it will allow or deny access to the network. Looking at each of the parts of Figure 12.4, one can explore the key roles in the 802.1x standard. The authentication server, authenticator, and supplicant are three main roles of any 802.1x exchange. They each perform specific roles in processing the authentication exchange and allowing correctly authenticated devices or users onto the network.

12.6.1 Authentication Server The authentication server provides the access granting and access rejecting features. It does this by receiving an access request from the authenticator. When the authentication server hears a request, it will validate it and return a message granting or rejecting access back to the authenticator. This is the back end of the 802.1x standard and, per the standard, the operation of this server is defined in another standard (i.e., RADIUS). © 2006 by Taylor & Francis Group, LLC


192


RADIUS

x

2.1

EA

0 /8

P

Authenticator Authentication server 802 1x from supplicant (Laptop) to authenticator (Access point)

Supplicant

RADIUS from authenticator (Access point) to authentication server (RADIUS server)

Figure 12.4 802.1x Overview.

12.6.2 Authenticator The authenticator is the first piece of network electronics that an 802.1x device will attempt connection. In this example, it is a wireless access point, although it can be anything providing access to the network. The device’s role is to let only EAP packets pass through and then wait for an answer from the authentication server. Once the authentication server responds with an accept or reject message, the authenticator acts appropriately. If the message is returned, it is a reject message and it will continue to block traffic until the result is an access accept. When the accept response comes from the authentication server, the authenticator then allows the supplicant the ability to access the network.

12.6.3 Supplicant The supplicant is the device that wants to connect to the 802.1x network. This can be a computer, laptop, PDA, or any other device with a network interface card. When the supplicant connects to the network, it must go through the authenticator. This authenticator only allows the supplicant to pass EAP request traffic destined for the authentication server. This EAP traffic is the user’s or device’s authentication credentials. Once the authentication server determines that the user or device is allowed on the network, it will send an access-granting message.




8 Octets

1 Octets 1 Octets

SNAP HEADER

Protocol Packet version type

2 Octets Packet body length

193

1-N Octets

EAP-Packet EAPOL-Start EAPOL-Logoff EAPOL-Key EAPOL-ASF-Alert

Packet body

0000 0000 0000 0001 0000 0010 0000 0011 0000 0100

Figure 12.5 EAPOL diagram.

12.6.4 Extensive Authentication Protocol over Local Area Network (EAPOL) Extensive Authentication Protocol over Local Area Network (EAPOL) is part of EAP, although it is outlined inside the 802.1x standard. Because of this, it is located in the 802.1x section in this book. This is because the 802.1x standard allows certain EAP message types to pass through an authenticator to the supplicant. To define what messages are allowed to pass through, each message type and frame format had to be included inside the 802.1x standard. The EAPOL standard calls out the process and frame structure used to send traffic from the authenticator to the supplicant. This traffic is outlined with six frame types. This means that only these six frame types are allowed to pass through an access point to a client. The IEEE created room for more, although the current standard only outlines six. Figure 12.5 shows each frame type, along with what value is used to identify the frame. EAPOL-Packet. This frame type is used to identify the packet as an EAP packet. EAPOL-Start. This frame type is used to begin an EAP conversation or an 802.1x authentication. EAPOL-Logoff. This EAPOL frame is used to end an EAP conversation or an 802.1x authentication. EAPOL-Key. This EAPOL frame is one of the most security-related frames. It is used to exchange keying information between the authenticator and the supplicant. EAPOL-Encapsulated-ASF-Alert. This is an EAPOL frame used to carry SNMP trap information out a non-802.1x authenticated port.



194


The most involved EAPOL type is the EAPOL-Key frame. This frame is used to send keying material, like dynamic WEP keys. The only key frame defined in the 802.1x standard is the RC4 WEP key. In 802.11i, one will see that some changes were made to the operation of the EAPOLKey frame to accommodate other encryption cipher types outside of RC4.

12.7 Remote Authentication Dial-In User Service (RADIUS) RADIUS is an acronym for Remote Authentication Dial-In User Service, a protocol used in network environments for authentication, authorization, and accounting. RADIUS can run across many types of devices, such as routers, servers, switches, modems, VPN concentrators, or any other type of RADIUS-compliant device. The protocol works by creating an encrypted tunnel between the network device and the RADIUS server. This tunnel is used for sending all the Authentication Authorization Accounting (AAA) information about who a user is, where the user is allowed to go, and where the user actually did go. To start this encrypted tunnel, a phrase or password called the shared secret is needed. The shared secret is located on the RADIUS participating network device and the RADIUS server. Once the shared secret is correctly set up, secure communication can take place. One of the benefits of RADIUS is the use of a common database of users to provide these AAA services across multiple device types. The database that RADIUS uses for storing usernames and passwords can be set up to point to many different types of directories. This means RADIUS can use most existing directory structures, such as Microsoft Active Directory (MS-AD), Novell Network Directory System (NDS), Lightweight Directory Access Protocol (LDAP), and many other common directory types. This protocol allows administrators to centrally locate and administer user access and accounting for all network equipment as well as remote access. RADIUS prevents many of the headaches associated with properly removing access to network equipment when employees are terminated. Once an organization has deployed RADIUS, user access could easily be removed in the event of a termination. This was unlike in the old days when administrators would have to manually change usernames and passwords on all network equipment. The RADIUS protocol specifications are currently defined in RFC 2865 and RFC 2866. RFC 2865 focuses on the access portion of RADIUS, allowing user access into devices or onto the network. RFC 2866 focuses on the accounting portion of RADIUS, allowing administrators to track changes




195

and access to network devices as well as general access to the network itself. The original RFC numbers were 2138 and 2139; these were updated to address a number of security-related concerns. Another major reason for this update was to change the UDP port number of RADIUS from the original port numbers of 1645 and 1646 to 1812 and 1813. The protocol was changed because the UDP port number of 1645 was already designated by IANA for the datametrics service and not for RADIUS. This is why most RADIUS servers support all four of these ports by default. Now that we have had a look into the primary usage and history of RADIUS, one can now look at how RADIUS relates to wireless. First, RADIUS could be used as an access method to administer the access point. This is similar to how it would be used to administer routers or switches. The access point and the RADIUS server would have a shared secret and that would be used to set up an encrypted channel that can carry user authentication traffic. Another approach is to use RADIUS, as explained in the 802.1x standard, as a back-end user authenticating mechanism. RADIUS itself provides this feature, so the 802.1x standard used that instead of creating its own authenticating mechanism. With this scenario, the access point would need to be set up correctly with the RADIUS server’s shared secret and the access point would keep track of the user’s request to enter the network. This means the user would only negotiate its authentication with the access point, not the RADIUS server. This is similar to how RADIUS would be set up on a remote access device. A user would ask to enter the network; the user would then be prompted by the network device to provide some type of authentication. Once the user provided authentication, the network equipment would verify it against the RADIUS server user database. If the credentials are correct, the user would be allowed onto the network; if the credentials are not correct, the user would be denied access. RADIUS has only four types of packets for authentication and there are other packet types for accounting. However, this section only focuses on the authentication packets. The four types are as follows: 1. Access-Request. This packet allows the RADIUS sequence to take place. 2. Access-Accept. This packet informs the RADIUS client that the authentication provided to it was correct. 3. Access-Reject. This packet informs the RADIUS client that the authentication provided to it was incorrect. 4. Access-Challenge. This packet is used to challenge a RADIUS client for its authentication credentials.



196


RADIUS packet formats Code

Identifier

Length

Authenticator

• Code • Identifier

1 Access-request

• Length

2 Access-accept

• Authenticator

3 Access-reject 11 Access-challenge

Figure 12.6 RADIUS frame format.

Looking more closely at the RADIUS packet in Figure 12.6, one can see that all four packet types consist of the same packet format. They are identified by the code field. This field has a number of available codes that could be used; however, the only ones addressed here are the AccessRequest identified by a 1, the Access-Accept identified by a 2, the AccessReject identified by a 3, and the Access-Challenge identified by a 1. The next field (the Identifier) is used to match requests and responses to each other. This ensures that multiple RADIUS conversations do not get mixed up as to what messages go to what device. The Length field is used to identify the length of the packet; because the RADIUS packet can have up 2000 attributes inside it, a mechanism is needed to measure the packet length. The last field, the Authenticator field, is the field in which the password is protected; this password is protected by a hashing mechanism. RADIUS has a lot of complexity. According to RFC 2865, each RADIUS packet can be up to 4096 bytes, allowing 2000 attributes into a single packet. Setting up RADIUS can be easy; however, depending on which vendor solution one decides on, there could be added complexity. Once thing to note is that there is no security complexity; time has shown that if security is too complex, it will be avoided or not installed. When looking at the RADIUS server, as well as any authentication servers for that matter, the details surrounding protecting the server itself are often forgotten. This stems from network people building and administrating networks — not servers. This leaves the server that RADIUS sits on exposed. For example, if the network people build and administrate networks, they most often do not have the skills to secure servers. One can have the most secure network in the world; however, if someone can easily hack into one’s authentication server, then one’s whole network is compromised.




197

12.8 Extensible Authentication Protocol (EAP) Extensible Authentication Protocol (EAP) is a standard method of performing authentication to gain access to a network. When Password Authentication Protocol (PAP) first came out, security issues quickly made it a less desirable authentication method. After that, the Challenge Handshake Authentication Protocol (CHAP) came out and this also quickly became plagued with security issues. The industry decided it was easier to make an authentication protocol act the same way no matter how or what type of authentication validation took place. This meant that for the first time a protocol could be inserted into products and software that allowed for passwords, tokens, or biometrics without having to write any extra code to support the different methods. This is how and why EAP was created. To use EAP, one must specify inside the type field what kind of authentication one is going to use. This allows one to use EAP for password, tokens, and other authentication types. EAP can adapt to security issues and changes by leveraging different methods of authentication. EAP also is able to address new and always-improving authentication techniques without having to make any changes to EAP supporting equipment. When EAP was created, a need for PPP compatibility was required. This helped ensure that a large base of equipment could handle EAP without major modifications. To get this compatibility, EAP was included as a Point-to-Point Protocol (PPP) type inside a PPP packet itself. This allowed for any device supporting PPP to be able to support EAP. EAP remained this way through RFC 2284. As EAP matured and required tighter integration with the 802.1x standard, its placement inside PPP was evaluated. The result of that evaluation is RFC 3748. Quoted below from RFC 3748 is the reasoning behind the changes to EAP that were needed to support the 802.1x standard. “The IEEE 802 encapsulation of EAP does not involve PPP, and IEEE 802.1X does not include support for link or network layer negotiations. As a result, within IEEE 802.1X it is not possible to negotiate non-EAP authentication mechanisms, such as PAP or CHAP [RFC1994].” EAP is defined in a PPP packet by the Hex code C227. PPP does not require authentication, although it does support it. As one can see from Figure 12.7, the EAP process starts by telling the device that the PPP authentication method is taking place and that method is EAP. Looking at the EAP portion of the packet itself, one can see that it is made up of four main fields: code, identifier, length, and data. These fields are used to identify what the packet function is, how long it is, and which current EAP negotiation the packet belongs to. Each of the sections in the EAP packet is identified in Figure 12.8, outlining each byte that makes up an EAP frame.



198


PPP Frame Type 3

Length 4

Authentication-Protocol C227

8 Bytes

8 Bytes

16 Bytes

EAP Frame Code

Identifier

Length

Type

Data

8 Bytes

8 Bytes

16 Bytes

8 Bytes

Variable length

Type

Data

Figure 12.7 PPP EAP frame format. EAP frame Code

Identifier

Length

8 Bytes

8 Bytes

16 Bytes

8 Bytes Variable length

Figure 12.8 EAP frame detail.

Code field. The Code field has a single octet and currently allows for four types of codes. The Request code makes an initial request; the Response code informs the requestor that their request was made. These codes are differentiated by the Code field being set to 1 or 2. The next set is the Success code, which validates that the authentication took place. It can inform the other party with a succeeded or failure code. These packets are distinguished from the other using a 3 in the Code field to indicate a success and a 4 to indicate a failure. When the RFC was updated to 3748, a note was placed in the Success section dealing with the arrived order of success packets. What this means is that if a client receives a Success frame before any Response frames during the EAP process, it will drop the frame. This was created to prevent attempts to circumvent the security of EAP. Identifier field. This section is also a single octet long. The function of the identifier attribute is to match requests to responses. This is used to distinguish when multiple clients authenticate with EAP at the same time. In this scenario, each client has a unique identifier that keeps its success or failure codes matched to each individual request. © 2006 by Taylor & Francis Group, LLC



199

Length field. This field is two octets long and indicates the length of the EAP packet. The length includes the code, identifier, length, type, and, of course, the actual data portion. Type field. The Type field has one octet and identifies the structure of an EAP Request or Response packet. This is where each EAP type allows for many kinds of authentication to take place. The EAP type is discussed later when we look at a number of the common EAP types that relate to wireless. This section has four main type codes to control the transmission of EAP credentials to the authentication server. These four types of type codes are identity, notification, NAK, and MD5-Challenge. The identity code is used to query the identity of a client and to start the process of identifying what type of EAP method will be used to authenticate. The notification code is used to notify the client of something or to get them to perform some type of action before continuing. The next code is NAK, which is valid only in response messages. This code is used for informing the client that the authentication type is not valid. There is also the ability to send an extended NAK, which details what EAP types the authenticator would accept. Data field. This section has the actual data that makes up the EAP request. Inside this part is where a user would provide the authentication portion. If EAP is using MD5, that would mean the user would have to submit a username and password to the access point. This information would be hashed and then located inside the data field of the EAP packet. One of the main points in using EAP is the ability to leverage multiple types of authentication mechanisms. This has helped EAP from becoming obsolete due to security vulnerabilities or protocol weaknesses. The ability to use multiple authentication types is located in the Type field on an EAP packet. The original standard as well as the new RFC 3748 only lists three main EAP types. These types are MD5 Challenge, One Time Password (OTP), and Generic Token Card (GTC). Today there are a number of different EAP types, some of which are vendor specific, some detailed in the EAP standard, and some detailed within their own standard documents. Others are industry standards on their own, detailed by IETF documents or RFC documents. In the subsections below, the most widely used wireless-related EAP types are detailed and examined.

12.8.1 EAP-MD5 Extensible Authentication Protocol-Message Digest version 5 (EAP-MD5) is one of the most limited EAP types included in the EAP RFC. This version © 2006 by Taylor & Francis Group, LLC


200


EAP request or start EAP identity EAP response with identity

RADIUS access request

EAP request

RADIUS access challenge

EAP response EAP success

RADIUS access accept

Figure 12.9 EAP-MD5 detail.

uses an MD hashing algorithm to validate user credentials. Some of the other types of EAP methods create encrypted tunnels and then inside these tunnels they perform EAP-MD5 validation. One of the requirements of EAP MD5 is a shared secret. This secret must be shared out of band so two parties know the same secret. That secret is then used to encrypt a challenge to verify that the other party has the same secret. Looking at Figure 12.9 one can see how EAP-MD5 works. The first step is to have the client send a request packet to the access point. Before that, of course, the shared secret must be exchanged out of band. Once the access point hears this request, it will respond, asking the client to provide its identity. The identity would state what EAP type they are both going to use. The RFC 3748 for EAP allows other identification items to be located in the EAP identity message, although the primary function of the EAP Identity field is to inform the authentication server of what type of EAP method the client will be using. Contrary to what the name implies, the Identity field should not be used for user identity — only for EAP type identity. Continuing with the example, the EAP type inside the EAP identity frame section would be set to MD5. Once the access point receives this identity code inside the frame of message 2, it would forward it to the authentication server. The authentication server would next send out a challenge to the client. This challenge is unencrypted and sent in cleartext. Once the client receives the challenge, it will use the shared secret to encrypt the challenge and send it back to the authentication server. Once the authentication server receives the challenge, it will perform the same operation that the client did. It will validate that the hash it has come up with matches the one that the client sent. If this is true, an EAP success message is sent; if for some reason something was not correct, then the authentication server would send a failure message with a reason code. This reason code would help to understand why the EAP frame was not accepted.




12.8.2 EAP-TLS The Extensible Authentication Protocol-Transport Layer Security (EAP-TLS) method is described in RFC 2716 and was created by Microsoft in October 1999. The RFC was built on the following RFCs: RFC 2284 for PPP and RFC 2246 for TLS. TLS came about from the older SSL protocol. Netscape created Secure Socket Layers (SSL) and used it for secure Web browsing. Once the Internet took off, updates to SSL were required. In 1996, the IEEE created TLS based on Netscape’s SSL and Microsoft’s Private Communications Technology (PCT). This EAP method uses certificates to authenticate users and requires certificates at both the server and client end. This is where TLS plays into this standard; TLS already created a good way to perform the needed certificate validation steps. This particular EAP method is one of the strongest, although it prevents usage unless one is accessing the network from a computer with one’s client certificate already installed. Setting up a wireless network with 802.1x and EAP-TLS requires some up-front work and planning. First, one must have a certificate authority (CA); this server will function as the distributor of both client and server certificates. Also needed would be an AAA server that supports EAP-TLS type. Finally, one needs a client that can support this EAP type. Once all the pieces are in place, the next challenge of correctly configuring each piece to interact is necessary. The EAP-TLS method works similar to the other EAP methods. To understand it, we present a high-level overview of the authentication process using this EAP method (Figure 12.10). It starts with a request packet originating from the client through the authenticator bound for the authentication server. Once this request is received, the authentication server will send back a response to the client asking for its identity. Next, the client will provide identity by requesting the use of EAP-TLS. Once this is received, the authentication server will send its public key certificate to the client. Once the client gets this, it will respond with its public key and a secure channel will be set up. This is very similar to how a secured Web page works, although in this case a client validation takes place. Now that both ends have authenticated to each other, another EAP process can take place inside the EAP-TLS tunnel to allow secure authentication to take place. For example, an EAP-MD5 exchange can take place inside the encrypted tunnel, allowing the weaker authentication to be secure because it is taking place inside an encrypted tunnel.

12.8.3 EAP-TTLS The Extensible Authentication Protocol-Tunnel Transport Layer Security (EAP-TTLS) is an IETF draft document created by Funk Software Inc. The most recent version is named draft-ietf-pppext-eap-ttls-05.txt and was © 2006 by Taylor & Francis Group, LLC


202


EAP request or start EAP identity EAP response with identity EAP request EAP type = TLS TLS start bit set

EAP response EAP type = TLS

RADIUS access request EAP Type = TLS RADIUS access challenge EAP type = TLS

TLS client_hello

RADIUS access request EAP type = TLS

EAP request EAP type = TLS

RADIUS access challenge EAP type = TLS

TLS server_hello, TLS certificate TLS server_key_exchange TLS certificate_request TLS server_hello_done

EAP response EAP type = TLS TLS certificate TLS client_key_exchange TLS certificate_verify TLS change_cipher_spec

EAP request EAP type = TLS TLS change_cipher_spec TLS finished


RADIUS access challenge EAP type = TLS

EAP response EAP type = TLS


EAP success


Figure 12.10 EAP-TLS detail.

created in July 2004. The reasoning behind creating a new EAP type was based on an opportunity Funk Software saw in the market. This opportunity was a need to support older devices that were not able to perform the new authentication types. This gave Funk the idea to write an EAP type that allowed for secure communication of credentials, along with the ability to allow legacy authentication types. This EAP type has helped Funk offer secure wireless solutions for older client equipment. The EAP-TTLS method (Figure 12.11) works by taking the authentication and protecting it inside a TLS tunnel; however, unlike EAP-TLS, EAPTTLS can use authentication types outside of EAP for the client authentication portion. Some older methods of authentication supported are PAP, CHAP, MS-CHAP, MS-CHAP-V2, and many others. Looking at how this works, one sees the main EAP process discussed thus far. The client makes a request with a request packet. This packet originates from the client through the authenticator bound for the authentication server. Once this request is received, the authentication server will send back a response to the client asking for its identity. Next, the client will provide identity © 2006 by Taylor & Francis Group, LLC



EAP request EAP identity EAP response with identity

RADIUS access request EAP type = TTLS

EAP request pass through

RADIUS access challenge EAP type = TTLS

EAP response EAP type = TTLS

RADIUS access request EAP-response pass through

TTLS client_hello

EAP request pass through

RADIUS access request EAP type = TTLS server_hello, certificate server_key_exchange server_hello-done

EAP response EAP type = TTLS client_key_exchange change_cipher_spec Finished

EAP success


RADIUS access challenge EAP type = TTLS change cipher spec finished

Now authentication method such as CHAP takes place inside TLS tunnel

Figure 12.11 EAP-TTLS detail.

by requesting the usage of EAP-TLS. Once this is received, the authentication server will send its public key certificate to the client. An optional configuration is to have the client send a certificate to validate that it is the correct client. Our example will not have the client using a certificate because the whole point of EAP-TTLS is to leverage old equipment, which cannot use advanced authentication methods such as Public Key Infrastructure (PKI). If one is going to use or already has PKI, EAP-TLS would be the better option. Once the client receives the certificate from the authentication server, it will use EAP-TTLS to set up a secure channel between him through the authenticator to the authentication sever. This is done using TLS and the public key of the authentication server. These are almost the same steps that any client would take in accessing a Web page that is protected by TLS or HTTPS. Once this is complete, a number of other standard types of authentication can take place inside this TLS tunnel. This is where EAP-TLS and EAP-TTLS defer in EAP-TLS all that could be done at this point is to perform another type of EAP such as EAP-MD5 inside the already set-up EAP channel. With EAP-TTLS you can use other forms of authentication that are not possible on device that do not support EAP. © 2006 by Taylor & Francis Group, LLC


204


EAP request or start EAP identity EAP response with identity LEAP requires user and password EAP request EAP type = LEAP EAP response EAP type = LEAP

RADIUS access request EAP type = LEAP RADIUS access challenge EAP type = LEAP RADIUS access request EAP type = LEAP

EAP success


LEAP challenge Used to authenticate RADIUS to client

RADIUS response with key

Figure 12.12 LEAP detail.

12.8.4 LEAP Lightweight Extensible Authentication Protocol (LEAP) is a Cisco Systems proprietary protocol. Cisco did release the source code for vendors that wanted to incorporate LEAP into their wireless adapters. The list of vendors includes D-Link, Dell, SMC, 3Com, and Apple. The code for LEAP is still considered Cisco Systems’ intellectual property and is available for use only under a non-disclosure agreement (NDA). LEAP works somewhat differently than the main EAP types discussed thus far (Figure 12.12). The EAP process starts out with the request from the client directed toward the access point. When the access point hears this, it will send the response and ask for identity. Most other EAP types just ask for the client EAP type; LEAP will respond with a username in the Identity field. Once the access point hears this, it will forward the packet to the authentication server or RADIUS in most wireless cases. The server will then send a challenge to the client through the access point. The challenge is made up of an 8-byte message. When the client hears this message, it will perform a number of steps in order to respond. First, it will use MD4 hashing, which will produce a 16-byte hash of the password. This hash will have the last 5 bytes padded with nulls. This will create a total 21-byte hash. This hash will be broken up into 7-byte chunks and each chunk will go through a DES encryption process. Each process will use the 7-byte chunks as a key and the original 8 bytes as the plaintext. This will result in a number of DES operations, creating a ciphertext that is 24 bytes long. This text will be sent to the authentication © 2006 by Taylor & Francis Group, LLC



server where it will do the same operation in reverse and derive the original password. Once it has this password, it will make sure that the passwords match that particular user. If a match is found, the authentication server will send an access-accept message to the access point. If not, the authentication server will send an access-reject message to the access point. Once the access point hears either of these messages, it will send an EAP failure or EAP success message to the client.

12.8.5 PEAP Protected Extensible Authentication Protocol (PEAP) was created as a joint effort between RAS, Microsoft, and Cisco Systems. Currently, PEAP is in an IETF draft called draft-josefsson-pppext-eap-tls-eap-08.txt, last updated in July 2004. Because of the fact that it still exists in draft form, updates may change its version number or document name. PEAP was a move by the industry to make a single EAP method that multiple vendors could share. The three vendors that created the standard implemented it each in their own way; this made Microsoft and Cisco versions of PEAP different and not interoperable. This has been slowly working itself out of the system, although do not expect an easy integration when trying to use PEAP methods from Microsoft and Cisco interchangeably. One of the main advantages of PEAP is the ability to have a strong EAP type that does not require client certificates, such as EAP-TLS. PEAP works similar to EAP-TLS by creating an encrypted tunnel with TLS and then performing another EAP method inside this encrypted tunnel. Unlike EAP-TLS, when PEAP performs this process, it does not validate a client certificate. This is where Cisco and Microsoft differ; each of them uses a different method after the TLS connection is created. The PEAP process (Figure 12.13) starts similar to the other EAP processes with a request packet sent to the access point. Once the access point hears this request, it will respond with a response packet. This response will also ask the client for its identity. The identity will be the EAP type; in this case, one will see EAP-Type=PEAP field in the frame. This will instruct the authentication server that the EAP method being used will be PEAP. The traffic will just pass through the access point, which can also be called the authenticator. Once the EAP method has been determined, a TLS tunnel is set up using the authentication server’s certificate to establish the TLS tunnel. Once a TLS tunnel is created, a new EAP process takes place inside the tunnel to authenticate the client. If one is using the Microsoft PEAP version, one will be using MSCHAPv2 as an EAP type. This EAP method is defined in the IETF document called draft-kamath-pppext-eap-mschapv2-01.txt, the latest version of which was posted in April 2004. © 2006 by Taylor & Francis Group, LLC


206


EAP request or start EAP identity EAP response with identity EAP request EAP type = PEAP V = 2 TLS start bit set EAP response EAP type = PEAP V = 2

RADIUS access request EAP type = PEAP V = 2 RADIUS access challenge EAP type = PEAP V = 2 RADIUS access request EAP type = PEAP V = 2

TLS client_hello

EAP request EAP type = PEAP V = 2

RADIUS access challenge EAP type = PEAP V = 2

TLS server_hello, TLS certificate TLS server_key_exchange TLS certificate_request TLS server_hello_done

EAP response EAP Type = PEAP V = 2

RADIUS access request EAP type = PEAP V = 2

TLS certificate TLS client_key_exchange TLS certificate_verify TLS change_cipher-spec

EAP request EAP type = PEAP V = 2 TLS change_cipher-spec TLS finished

EAP response EAP-TVL identity

RADIUS access challenge EAP type = PEAP V = 2 RADIUS access request EAP-TVL type

EAP request EAP-TVL EAP success

RADIUS access accepet

Figure 12.13 PEAP.

12.8.6 EAP-FAST Extensible Authentication Protocol-Flexible Authentication via Secure Tunneling (EAP-FAST) is an IETF document created by Cisco Systems in February 2004. The current document is named draft-cam-winget-eap-fast00.txt and is located on the IETF Web site. Cisco had some security flaws released on their LEAP method and instead of fixing them, they abandoned the proprietary standard and created this EAP method. This EAP method supports a fast roaming time compared to the other EAP standards. This timing was a critical requirement for many companies that have Wi-Fi phones or some time-sensitive application. This made Cisco Systems in need of a fast, secure EAP method for wireless authentication. © 2006 by Taylor & Francis Group, LLC



EAP request EAP identity EAP response with identity EAP request EAP type = EAP-FAST EAP-FAST start, S bit set, A-ID

EAP response EAP type = EAP-FAST TTLS client_hello with PAC opaque

EAP request pass through EAP response EAP type = EAP-FAST TLS change_cipher_spec TLS finished

RADIUS access request EAP-response pass through RADIUS access request EAP type = EAP-fast TLS server_hello, TLS change_cipher_spec, TLS finished


EAP request

RADIUS access challenge EAP type = EAP-fast change cipher spec finished

EAP response user identity


EAP success


Figure 12.14 EAP-FAST.

The EAP-FAST method (Figure 12.14) works by having what is called a protected access credential (PAC), which is similar to a certificate. The PAC is a 32-octet key that is used to set up an encrypted tunnel between the client and the authentication server. One unique thing about this key is its ability to have the keys dynamically allocated. This PAC dynamic allocation process is called auto-provisioning, phase 0, or in-band provisioning. Whatever one decides to call it, it works by setting up a TLS connection and passing MSCHAP V2 credentials to access the network. To note is that the TLS session is only authenticated by the server’s public key. After the initial network connection is made, a new PAC can be sent down this encrypted channel. Cisco does not recommend this approach if avoidable; however, it does allow for an easy upgrade from LEAP to EAP-FAST. The other approach is to use out-of-band key exchange or good old sneaker net. To do this, just key in the same key at both the user end and the server end. To get around any key management problems, the PAC has the ability to change at specified intervals once a secure connection has been established. Looking more closely at the PAC, one can see that it is made up of three main portions: (1) the master key, (2) the opaque, and (3) the PAC © 2006 by Taylor & Francis Group, LLC


208


info. These three parts each have their own functions in supporting EAPFAST. The master key is generated randomly by the authentication server; this generation is done to produce strong entropy. After the key is generated, it is sent to the client. This key has the authority ID bound to it as well. The second portion of the PAC looks at the opaque, the authority ID, and key lifetime that are encrypted by the client’s master key. The final portion of the PAC is the info section, which has the authority ID and can have the key lifetime. This field is used to difference multiple PACs on a client to support connection to multiple networks using multiple sets of credentials. Now that the PAC has been created or keyed in, the EAP process can start. As with the other EAP processes, this one starts in the same manner with the request packet from the client. The access point answers the request and a response is sent back requesting identity. The client then sends its identity, which is not the client’s identity but the actual EAP type identity. This identity request just requests the usage of EAP-FAST. Once the access point hears this identity request, it will respond with an EAP request to send the client’s PAC. This message with the PAC will be passed on to the authentication server. The authentication server will then send an Authority ID, which is a field inside the PAC. This field identifies the correct PAC from a client PAC list. Once the client hears the authority ID and matches that against the correct PAC, it can start to create a TLS tunnel based on the Master key for that PAC. This will create a tunnel that can be used to perform further authentication.

12.9 Wi-Fi Protected Access (WPA) Wi-Fi Protected Access (WPA) has an interesting history in relation to how it became a standard. When the security of WEP was broken, the industry turned to the IEEE to fix it. The IEEE said it would create the 802.11i wireless security standard. This standard dragged on and was very slowmoving. As it took longer and longer to ratify, wir eless device sales declined. This decline was due to the lack of a standard secure wireless networking method. With this all-so-needed standard lacking, the wireless manufacturers started to push the IEEE and other standard boards to ratify something so they could produce secure standard products. With the pushback of the 802.11i release date, the Wi-Fi Alliance decided that it would create a subset 802.11i standard called WPA. The Wi-Fi Alliance created WPA by leveraging what the 802.11i task group had already done and formalized it into WPA. This meant that any major changes to the 802.11i standard would influence future versions of WPA. This was seen with WPA and WPA2. Today, with 802.11i complete, the use of WPA has greatly decreased. © 2006 by Taylor & Francis Group, LLC



The WPA standard supports two methods of authentication and key management. The first one is EAP authentication with the 802.1x standard. This method works through the use of the 802.1x protocol and a back-end authentication server. It leverages EAP for in-air authentication and RADIUS for back-end authentication. This method is the more secure of the two and provides the lowest amount of end-client administration. The next available option is to use preshared keys. This option requires that a key be applied to the devices and the wireless access points. This also means that everything has the same password entered. To combat someone using this key to eavesdrop on others’ conversations, WPA uses a method that creates a unique session key for each device. This is done by having a preshared key called the group master key (GMK) that drives a pair transient key (PTK). How this works is explained in the section on 802.11i. This second solution was added to WPA for home and small office support. In a house or small office, one is unlikely to have an authentication server such as RADIUS. A PSK is a 256-bit number or a passphrase that is 8 to 63 bytes long. WPA does support TKIP and MIC for older devices. It will also perform AES, although the method it uses is a little different from the one defined in 802.11i. It has most of the features addressed in the 802.11i section. Some of these include the ability to negotiate a cipher suite or authentication method with robust security network (RSN) information elements. One of the reasons why 802.11i was not ratified was because of certain requirements that were not well defined at the time. With the WPA standard using whatever the IEEE 802.11i task force had already completed, some changes were needed to be able create this interim standard. These changes led to a number of differences between the two standards. The first big difference is that WPA supports TKIP by default. This is unlike 802.11i, which supports AES CCMP by default. The next item is the fact that WPA does not even support AES CCMP; it does, however, support AES, just not in the variation that 802.11i calls out. The final major item that differentiates WPA from 802.11i is the RSN IE. This is used to pass the supported cipher settings between the wireless access points and clients. In 802.11i, this portion was not well defined so the WPA standard had to create some newer rules without having them affect anything that might be done to the RSN IE from the 802.11i task group. This was accomplished by creating a WPA IE and using different values to distinguish them from one another. This helped so that once the RSN IE was well defined, it was not difficult to put it into WPA.

12.10 802.11i The recently ratified 802.11i security standard came about in response to a need to improve the security of 802.11 networks to a level sufficient to © 2006 by Taylor & Francis Group, LLC


210


warrant wireless as a generally accepted secure transport medium. In this standard, the IEEE outlined a secure way to access wireless networks. It also tried to mitigate the now-enormous amount of threats that were making wireless networking a real risk for companies. Looking at the process of how 802.11i became a standard, one needs to go back to July 1999 when there was interest in enhancing the MAC layer of 802.11 for quality of services (QoS) and privacy. This built up enough steam to create the task group TGe in March 2000. After a year, it was determined that this group needed to split into a security group and a QoS group because of the large workload. This split created the TGi security working group. This group created the 802.11i standard and put it up for approval by voting on it. To proceed to the next level of voting, the working group had to have a 75 percent approval vote. This approval took a number of drafts over a three-year period. Once this was done, there was a level of sponsor ballots and finally the standard board’s approval process. On June 24, 2004, the standard body finally approved the 802.11i standard. Looking at 802.11i up close, one notices that it uses a number of standards, protocols, and ciphers, which have already been defined outside the 802.11i. A number of standards are also defined inside it as well; RADIUS, 802.1x, EAP, AES, RSN, TKIP, and others are some of the defined standards that are part of the 802.11i standard. Some of these are defined inside their own document and some of them are officially created inside the 802.11i document. The Robust Security Network (RSN) standard is used for dynamic negotiation of authentication and encryption. It is used to negotiate what kind of encryption a client can support as well as what type of encryption is required based on a policy. A deeper look into RSN is available in Section 12.10.1. Before we get into this, we need to address 802.11i standard as a whole. Another piece of the 802.11i standard is the ability to use EAP. It was determined that the 802.11i standard would not specify an authentication method or type; rather, it would allow a protocol that can perform multiple types of authentication inside itself. This is exactly what EAP does; it allows the use of many different authentication types from passwords, smart cards, certificates, and many others based on the same request, accept, and reject methods. For EAP to work correctly with the 802.11i standard, another well-known standard must facilitate the transmission of EAP between untrusted and trusted entities. This is where the 802.1x standard fits in; its main goal is to provide a framework for strong authentication and key management. The 802.1x protocol allows the access point only to permit an EAP request into the network. This is the case




until the client is properly authenticated. Once authenticated, key negotiation and subsequently network access can be achieved. As included in WPA, the 802.11i standard needed an option for environments in which an authentication server was not financially feasible. This authentication server was a requirement of the 802.1x standard. To make 802.11i viable for both large enterprises and small office/home office (SOHO) users, another method was necessary. This is where the preshared key method originated. This is very similar to WPA and its preshared key method. When a preshared key is used, each client uses a secret to create subsequence-keying material. This master key is the same across the network, just like WEP, although it is used to create a session-based key for each client. Before getting into the details of how 802.11i works, one needs to understand all the components that make up 802.11i so one is comfortable with a full system overview. Below is each main portion of the 802.11i standard. As said above, most of the standards and protocols inside 802.11i are either their own standard or are located inside the 802.11i standard. The subsections below outline the standards that are located inside the 802.11i standard.

12.10.1 Robust Secure Network (RSN) Robust Secure Network (RSN) was created as part of the 802.11i security standard. RSN specifies user authentication through IEEE 802.1x and data encryption through the Temporal Key Integrity Protocol (TKIP) or Counter Mode with CBC-MAC Protocol (CCMP). RSN also has the option to use TSN to allow for the use of older security methods such as WEP. TSN is explained in detail in Section 12.10.1.1. RSN uses TKIP and AES as encryption methods to protect the confidentiality of data. The TKIP solution is used for backwards compatibility for legacy devices, and the AES is what RSN is using as a long-term encryption method. AES is set up in Counter Mode with the CCMP. AES can be set up and used in multiple ways, so 802.11i stated that AES must be used in a method called CCMP. The RSN protocol also uses EAPOL-Key messages for key management. The description below reveals how RSN works with 802.11i in aiding to choose an available authentication method and encryption cipher scheme. Advertising the cipher suites supported on an access point and client is done through Robust Secure Network Association (RNSA) messages. These messages spell out the supported ciphers of each party and negotiate what method will be used to connect securely. These messages are located inside what is called an RSN IE or an RSN information element. A RSN IE is used to tell the other devices about what cipher suites the sending



212


Element ID

Length

Version

Group cipher suite

Pairwise cipher suite

Pairwise cipher suite list

AKM suite count

AKM suite list

RSN capabilities

PMKID count

PMKID list

1 octet

1 octet

2 octets

4 octets

2 octets

4-N octets

2 octets

4-N octets

2 octets

2 octets

16 octets

Figure 12.15 RSN-IE frame detail.

device supports. The RSN IE can be sent in a beacon from an access point or in an association request from a client. After an association request, a response will be returned with an RSN IE listing what requesting method matched the method supported by the other party. The standard allows RSN IE optionally to be inside each of the following management frame types:

Beacon Association Request Reassociation Request Probe Response

Looking at Figure 12.15, one can see the many parts that make up an RSN IE frame. Of the 11 sections of this frame, only the first three are required in all RSN IE transmissions. After the three required fields, all other fields must have the preceding field inserted with data or the frame will not be properly received. This means that if one needs the value in the ninth field, one must include all eight preceding fields. Element ID. This field supports 48 decimals or 30 hexadecimal digits. Currently, the only allocated element ID is 48, which stands for RSN. Length. The Length field identifies the total length of the RSN IE frame. Currently, the frame is only 255 octets long. When using a large number of cipher suites, one may run into a case where one can only support a limited number of cipher suites. This is due to the limit in the total size of the RSN IE frame. Version. The Version field is used to show what version of RSN is currently being used. It holds up to two octets. Today, only a single version of RSN exists. This version is shown as 1 in this field; Versions 0 and 2 are reserved for other versions. Group Cipher Suite. This is the cipher suite used to protect broadcast and multicast traffic. It holds up to four octets of information about the total group cipher suites used: up to two octets of information about the multicast and two octets for broadcast cipher suites used. Pairwise Cipher Suite Count. This field lists the number of selected pairwise cipher suites. It is only two octets long.




Pairwise Cipher Suite List. This field contains all the ciphers that were selected for the pairwise key. Each of the cipher suites is accounted for in the Pairwise Cipher Suite Count field. Each one has a corresponding type inside this field. Each one of the used cipher suites is four octets long. Of the four octets, three are used for the Organizationally Unique Identifier (OUI) field and a single octet is left to identify the cipher suite. AKM Suite Count. The Authentication and Key Management (AKM) Suite Count is used to determine how many different key management options are available, such as preshared keys or ones dynamically allocated with 802.1x. To note is that in an Independent Basic Service Set (IBSS), only a single AKM suite can exist. This field has a maximum size of two octets. AKM Suite List. The Authentication and Key Management Suite List is used to specify what key management options are available. Depending on the AKM Count Suite field, this field could have multiple, four-octet sections defining each key management option. Currently, only two options exist: pre-shared keys and 802.1X key management. This leaves a number of reserved and vendor-specific options to include later. RSN Capabilities. This is a two-octet field used to identify what RSN capabilities are available on the network. It identifies whether or not the device is capable of a pairwise key. It also allows the device receiving the RSN IE to understand if it can or cannot support RSN altogether; if it cannot support RSN, TSN can be tried. In the event that it is not understood, it is assumed that it cannot support RSN. PMKID Count. The Pairwise Master Key Identifier Count field is a two-octet field used only with reassociation. It is used to cache keys so that when a client is roaming, it does not have to go through the entire authentication process with each access point. This is to speed up the timing and lower the bandwidth as a client roams from one access point to another. The count is to define how many of these credentials are currently inside the RSN IE frame. PMKID List. The Pairwise Master Key Identifier List field is four octets for every PMKID identified with the PMKID Count field. This is where each of the different types of PMKID is stored. Currently, there are three main PMKID types listed. The first one is a cached PMK that has been obtained through pre-authentication with another AP. The second one is a cached PMK from an EAP authentication. And the third is a cached PMK from a PSK.



214


Table 12.3 OUI

3 octets OUI

RSN Cipher Suite Frame Detail Suite Type

1 octet Suite Type

Meaning

00-0F-AC

0

Use group cipher

00-0F-AC

1

WEP-40

00-0F-AC

2

TKIP

00-0F-AC

3

Reserved

00-0F-AC

4

CCMP

00-0F-AC

5

WEP-104

00-0F-AC

6-255

Reserved

Vendor OUI

Other

Vendor-specific

Other

Any

Reserved

Now let us look more closely at the different types of cipher suites. In RSN, one is only concerned about telling the other party what cipher suites are supported. Once that has been accomplished, one can decide if it is possible to negotiate a common cipher suite or method between both parties. In the RSN Cipher Suite frame section, there are six specified cipher suites and a number of reserved and vendor-specific cipher suites that can be used in the future. Today, the six supported cipher suites are identified with the hexadecimal code 00:0F:AC. Looking at Table 12.3, one can see that each of the following cipher suites is listed as 00:0F:AC and numbers 1 through 6. The RSN standard is a method to negotiate what types of security methods each client and each access point supports. These security methods are identified as cipher suites inside the RSN IE frame. These cipher suites allow for the use or non-use of any combination of security methods. This means that a policy could be put into place that negates the use of weaker security methods such as WEP and allows for a choice of TKIP or AES. This gives the architects or designers the ability to create a policy allowing or denying whatever cipher suites they might feel are weak or not needed.

12.10.1.1 Transition Secure Network (TSN) The Transition Secure Network (TSN) is part of the RSN portion of 802.11i. It is used to achieve backwards compatibility with older wireless systems. © 2006 by Taylor & Francis Group, LLC



It was carved out of RSN to provide this backwards compatibility. With RSN, as stated above, it is possible to have a number of authentication and encryption types running on an access point. To make sure that some of the weaker authentication and encryption types were not set up in RSN, they were taken out and considered TSN. This makes RSN more secure and allows an easy way to turn off all the older weak methods. With RSN, if one chooses not to support TSN, WEP will not be included as an option to negotiate between the access point and the wireless client.

12.10.2 Temporal Key Integrity Protocol (TKIP) The Temporal key Integrity Protocol (TKIP) was an interim solution developed to fix the key reuse problem of WEP. It later became part of the 802.11i and subsequently part of WPA standards. This meant there were various flavors of TKIP until 802.11i was finalized. One of the first notations about the theory and concepts of TKIP was published in December 20, 2001, by Russ Housley and Doug Whiting, in an article entitled “Temporal Key Hash.” This article described the general principle of TKIP, although it was not enough on which to base a standard. That is where 802.11i came in with a more in-depth creation of TKIP. TKIP was included in the 802.11i standards for backwards compatibility. The 802.11i standard did not want to use a cipher based RC4, so they chose AES (see Section 12.10.3). TKIP was put into 802.11i for the sole reason of helping older devices transition to 802.11i. To do this, 802.11i needed to support a protocol that could easily upgrade WEP to something safe enough to include in 802.11i. One of the main reasons for using TKIP over WEP came from the increased security and increasing number of attacks that were plaguing the WEP protocol. Using TKIP protected against these attacks and reduced the overall risk of operating a wireless network. The TKIP standard also saw value in the industry because the migration from WEP to TKIP was an easy one. In most cases, moving from WEP to TKIP involved a small firmware change. This meant that no hardware was required to make the change and also that most older, already purchased equipment would be able to upgrade to TKIP. Another interesting note about TKIP comes from Cisco Systems. Cisco came up with a TKIP solution well before the 802.11i standard defined one. This has led some people to wonder about which version of TKIP is on a certain product. Vendors other than Cisco also created TKIP-based solutions before the standard was ratified. Today, Cisco differentiates its versions of TKIP and the standard one by calling it the Cisco Key Integrity Protocol (CKIP). In Cisco products, one can specify to use TKIP, which is the 802.11i-compliant version, or CKIP, which is the Cisco-created version. © 2006 by Taylor & Francis Group, LLC


216


The TKIP encryption portion works in a two-phase process. The first phase generates a session key from a temporal key, TKIP sequence counter (TSC), and the transmitter’s MAC address. The temporal key is made up of a 128-bit value similar to the base WEP key value. The TKIP sequence counter (TSC) is made up of the source address (SA), destination address (DA), priority, and the payload or data. Once this phase is completed, a value called the TKIP-mixed transmit address and key (TTAK) is created. This value is used as a session-based WEP key in the second phase. In the second phase, the TTAK and the IV are used to produce a key that encrypts the data. This is similar to how WEP is processed. In WEP the first 24 bits of the IV are added in front of the WEP key and then used to create an encryption key that is applied to the data. Then the IV is inserted into the packet header. TKIP extended the IV space, allowing for an extended IV field, which holds an additional 24 bits. In the second phase, the first 24 bits are filled with the first 24 bits of the TTAK. The next 24 bits are filled with the unused portion of the TSC. This is safer than WEP because the key is using a different value, depending on who one is talking to. In WEP, each client or access point creates the same random value. Some products never even created a random value and just incremented the value by one, making it an easy target for hackers. The basis of TKIP came from the WEP protocol. In the 802.11i standard, TKIP is referred to as a cipher suite enhancing the WEP protocol on preRSNA hardware. This is espoused because RC4 is still used as a cipher, although the technique in which it is used has improved greatly.

12.10.2.1 TKIP Message Integrity Check (MIC) Similar to TKIP, the Message Integrity Check (MIC) had also many versions before 802.11i defined it as a single standard. Once this was done, MIC became known as Michael although the acronym MIC still remains. Today with 802.11i, ratified MIC is Michael and vice versa. The protocol itself was created to help fight against the many message modification attacks that were prevalent in the WEP protocol. The IEEE 802.11i standard describes the need for MIC in the following quote: “Flaws in the IEEE 802.11 WEP design cause it to fail to meet its goal of protecting data traffic content from casual eavesdroppers. Among the most significant WEP flaws is the lack of a mechanism to defeat message forgeries and other active attacks. To defend against active attacks, TKIP includes a MIC, named Michael.” The MIC was created as a more secure method of handling integrity checking compared to the IVC in WEP. The MIC is a hash that is calculated on a per-packet basis. This means a single MIC hash could span multiple frames and handle fragmentation. The MIC is also on a per-sender, per-receiver basis. This means that any © 2006 by Taylor & Francis Group, LLC



given conversation has a MIC flowing from sender A to receiver B and a separate MIC flowing from sender B to receiver A. The MIC is based on seed value, destination MAC, source MAC, priority, and payload. Unlike IC, MIC uses a hashing algorithm to stamp the packet, giving an attacker a much smaller chance to modify a packet and have it still pass the MIC. The seed value is similar to the WEP protocol’s IV. TKIP and MIC use the same IV space, although they have added an additional four octets to it. This was done to make the threat of using the same IV twice in a short time period less likely. The MIC is also encrypted inside the data portion, which means it is not obtainable through a hacker’s wireless sniffer. To add to this, the TKIP also left the WEP IVC process, which then adds a second, less secure method of integrity checking on the entire frame. To combat message modification attacks, the TKIP and MIC went a step further and introduced the TKIP countermeasures procedures. This is a mechanism designed to protect against modification attacks. It works by having an access point shut down its communications if two MIC failures occur in 60 seconds. In this event, the access point would shut down for 60 seconds. When it comes back up, it would require that all clients trying to reconnect change their keys and undergo a re-keying. Some vendors allow one to define these thresholds, although the MIC standard calls out these values. To prevent noise from triggering a TKIP countermeasure procedure, the MIC validation process is performed after a number of other validations. The validations performed before the MIC countermeasure validation are the frame check sum (FCS), integrity check sum (ICV), and TKIP sequence counter (TSC). If noise was to interfere with the packet and modify it, one of these other checks would be able to find it first, thus preventing the frame from incrementing the MIC countermeasure counter.

12.10.3 Advanced Encryption Standard (AES) One can apply AES in many different ways. The way that the 802.11i standard has chosen to apply AES is with CCMP, which is based on CBCMAC. CCMP was chosen for data integrity and authentication, with the Message Authentication Code (MAC) providing the same functionality as the Message Integrity Check (MIC) used for TKIP. Before diving into CCMP, one needs to look at AES and some of its modes. The first term is CTR; this is AES in Counter mode. This mode is used for confidentiality. The next mode is called CBC-MAC, which stands for Cipher Block Chaining Message Authentication mode. This mode is used for integrity. AES also has combined CTR and CBC-MAC to create CCM. CCM is the acronym for CTR/CBC-MAC mode of AES that incorporates both the confidentially of CTR and the integrity of CBC-MAC. © 2006 by Taylor & Francis Group, LLC


218


12.10.4 802.11i System Overview Having looked at each part that makes up the 802.11i standard, one can now look at it as a whole. One will see how the client connects to the access point, authenticates, and negotiates keys. Each one of these steps leverages the outlined standards discussed thus far. The client would first need to make a connection to the access point. This would happen through the normal open key authentication process (Section 12.2). Contrary to most 802.11 standards, 802.11i only allows for open system authentication. This is due to the discovery of a security flaw in shared key authentication. This flaw is detailed in Chapter 13. After the initial connection request, the client would need to hear an RSN IE broadcast or send a probe request with an RSN IE. Whichever way this RSN IE frame is sent, both clients and access points need to negotiate on a cipher suite for use. After sending the RSN IE frames and reaching a negotiation, the EAP process starts. This can start with the access point sending an EAP identity request or a client sending an EAPOLStart frame. Once the EAP process has started, it will go through the EAP authentication process associated with each particular EAP type. This process is outlined in Section 12.8. It ends with the client receiving an EAP success message from the access point. During this process, an AAA key is sent from the authentication server to the wireless end device. This key is used as a seed key to create the keys outlined below. The key exchange process takes the original 802.1x EAPOL-Key frame and makes some modifications, allowing for the use of WEP-40, WEP104, TKIP, and CCMP cipher suites. From the 802.1x section, the EAPOLKey frame only supports WEP-40 and WEP-104 keys. The 802.11i standard modified this and added the ability for the frame to carry TKIP and CCMP keys as well. A process known as the four-way handshake accomplishes this key exchange. This process takes two main keys and creates unique group and session keys for each client. These session and group keys are created from the two main keys: (1) the pairwise key or the pairwise master key (PMK) and (2) the group key or the group master key (GMK). In an 802.1x 802.11i setup, the PMK comes from the authentication server. If the 802.11i setup is using preshared keys, then the PMK is mapped to a password. The PMK is divided into three keys. The first key is the EAPOL-key confirmation key (KCK), which is used to provide data origin authenticity. The second key created from the PMK is the EAPOLkey encryption key (KEK), which is used to provide confidentiality. The last key is called the pairwise temporal key (PTK) and this key is also used for data confidentiality. To create the PTK, a pseudorandom function takes place with the access point’s MAC address, client MAC address, and a nonce sent from each side as well. This allows a single master key to




create multiple session keys without having to re-exchange a new master key each time. The next key with regard to 802.11i main keys is the group key or group master key (GMK). This key is similar to the PMK except that it is used for beacon and management traffic encryption. The same process of hashing senders’ and receivers’ MAC addresses and nonces is used to create a group temporal key (GTK) from a group master key. Having discussed the keys and how they are split up to accommodate session encryption, one can now look more closely at the four-way handshake. This handshake starts with the authenticator sending the supplicant a nonce. This is often referred to as the ANonce in the 802.11i standard. This nonce is a random value used to prevent replay attacks. This means that old nonces cannot be reused. After each party receives a message, the first step before any other is to check and see if the nonce was changed or if the same nonce was incorrectly reused. Once the wireless client receives the first message, it will check the nonce and then generate an SNonce. This nonce will be used in the next step to calculate the pair transient key (PTK). After the PTK is created, the client will then send the SNonce as well as the security parameters outlined in the RSN IE frame to the access point. This information is the second message in the four-way handshake. All of this information will be encrypted using the KCK, which will protect it from any modification while in transit. Once the access point receives this, it will check that the nonce is not an old value. Once this is done, it will also generate the PTK from the SNonce and ANonce, and then check the KCK to make sure it was not modified in transit. Once this is done, the third message in the four-way handshake will take place. This message is used to tell the client to install the PTK key that was created and, if used, this message will send a GTK to the client to install. Once the client receives this, it will check the KCK and, if it is correct, install the key or keys. The last message is a confirmation used to let the authenticator know that the client has successfully installed the keys and is ready to communicate using them. Figure 12.16 shows this four-way handshake as it takes place.

12.11 Wi-Fi Protected Access (WPA2) After 802.11i came out, the Wi-Fi Alliance wanted to continue the initial investment made in WPA. This created an issue because the 802.11i standard was now out and another standard was not what the industry needed. To keep WPA going, the Alliance decided it would go back to the core benefit the organization provided — standard interoperability testing and certification. In creating WPA2, the Wi-Fi Alliance made this



220


Generate random S-nonce

Generate random A-nonce Message 1: Nonce exchange (EAPOL-key message)

Calulate PTK using A-nonce and S-nonce Message 2: Send nonce, MIC (EAPOL-key message Calulate PTK using A-nonce and S-nonce Message 3: Nonce, MIC, RSN, IE, key (EAPOL-key message) Message 4: Acknowledgement MIC (EAPOL-key message) Set temporal encryption key and MIC

Set temporal encryption key and MIC

Figure 12.16 802.11i Four-way handshake.

version of WPA an interoperability mark similar to Wi-Fi. This mark ensures that any product carrying it has an interoperable 802.11i standard.

12.12 WLAN Authentication and Privacy Infrastructure (WAPI) China has decided that 802.11i is taking too long and they are better off creating their own standard. Even with 802.11i out today, China has still said they want to take their own approach. This has led to the creation of the WLAN Authentication and Privacy Infrastructure (WAPI) standard. This standard has many similarities to 802.11i, such as RADIUS and 802.1x. Getting any information about this standard is a violation of Chinese national security under China’s State Council Directive 237, which regulates commercial encryption ciphers and requires encryption technology to be developed and sold under a blanket of secrecy. To sell WLAN products within China, foreign equipment vendors must license WAPI through an agreement with one of 24 Chinese companies granted rights by the Chinese government. This has been a hard sell to




many industry leaders such as Broadcom, Texas Instruments, Intersil, Atheros, and many others. The Chinese government has asked them to provide all the intellectual property involved in their chipset. This intellectual property has made them a market leader; and to sell in China, they could lose that. As one might guess, many of these companies are not taking the risk.

12.13 Rogue Access Points Detection Detecting rogue access points has become almost an industry of it own. Many products came to market with the sole purpose of identifying rogue access points. When other companies saw these moves, they tried to adapt their current products to perform this function as well. Today we see a number of products that perform rogue detection from access points with built-in detection, to appliances, to stand-alone devices whose sole purpose is rogue detection. In the chapter on risks and threats (Chapter 2), there was very little discussion of rogue access points, merely what they are and how they pose a risk to the enterprise. This section discusses how to detect rogue access points. Each company that developed rogue access point detection software took a unique approach in doing so. This has given us a wide variety of techniques to use in detecting and combating rogue access points. Some products, such as Cisco’s wireless LAN solutions engine (WLSE), have the ability to detect and prevent rogue access points from connecting to the network. How WLSE works is by using other access points, which are managed by WLSE, as sensors. These access points listen for beacons or data frames coming from other access points that are not defined in the WLSE console. When this happens, the unknown access points are labeled as rogue. The next step, which is where Cisco has differentiated itself from other rogue detection systems, is the ability to leverage the wired network to shut down rogue access points. This is done by querying the Cisco switches in an attempt to find the rogue access point’s MAC address. This MAC address is easily read from the airwaves so WLSE already knows what it is looking for. When the query is made to the network, the switch, which has the access point’s MAC address in its Content Addressable Memory (CAM) table, will respond by saying that “I have that MAC address.” WLSE can then tell the switch to shut down that port and thus disconnects the access point’s connection to the network. This method can fight a moving rogue access point very well. As soon as someone moves the access point somewhere else, the whole process starts over finding, identifying, and then shutting down the rogue access point.



222


Some of the other approaches that have been made to go beyond the location and identification of rogue access points have to do with using denial-of-service attacks. This feature is interesting and might be considered illegal. This is definitely something that someone in a multi-tenant location may want to consider before using. This approach works on the basis of finding a rogue access point and then using nearby access points to send disassociation messages to all clients connecting to the rogue access point. As stated, in a multi-tenant location, this could mean disrupting a neighbor’s network. This could become a legal issue quickly. So far, no court cases have come out of the woodwork with regard to this particular issue, although it is just a matter of time.

12.14 Chapter 12 Review Questions 1. Which encryption mode was selected for 802.11i? a. RC4 b. AES c. AES-CCMP d. 3DES 2. Which EAP type is part of the EAP RFC? a. LEAP b. EAP-MD5 c. EAP-TTLS d. EAP-TLS 3. What EAP method did Microsoft, Cisco, and RSA create? a. PEAP b. LEAP c. EAP d. EAP-TLS 4. Does China have its own wireless security standard? a. Yes b. No 5. In a. b. c. d.

802.1x, what is the RADIUS server often called? Supplicant Authenticator Authentication server RADIUS server




6. In a. b. c. d.

802.11i, what mode is one using if WEP is enabled? RSN TSN TKIP CCMP

7. Which IEEE standards deal with security? (Select two) a. 802.11 b. 802.11b c. 802.11i d. 802.1x e. 802.11g f. 802.16 8. During the time frame when WPA was released and 802.11i was still being finalized, what reservations would an IT manager have about using TKIP? a. Weak keys b. Lack of interoperability c. Possibility of repay attacks d. Short age 9. Which EAP type is used to create a tunnel in which other, older authentication methods can take place? a. PEAP b. EAP-MD5 c. EAP-TTLS d. EAP-TLS 10. Which of the following is not a VPN technology term? a. IPSec b. SSL c. ESP d. AH 11. Which of the following provides layer 2 security by allowing for unique and changing encryption keys? a. 802.1x/EAP b. WEP c. TKIP d. AES



224


12. During an EAP negotiation, what happens after the identity response packet is sent from the client to the access point? a. The RADIUS server sends out a challenge. b. The access point sends out a challenge. c. The access point forwards the message to the RADIUS server. d. The access point converts the message and sends it to the client. 13. What is a new federal encryption standard that is used in a certain mode on wireless networks? a. RC4 b. WEP c. DES d. AES 14. What is used to ensure the integrity of packets traveling across the airwaves? a. Authentication server b. RADIUS c. MIC d. IV 15. WEP keys can also be thought of as __________. a. Tokens b. Secure IDs c. Key cards d. Shared secrets 16. Most wireless equipment is set to which better authentication method? a. Open key authentication b. Shared key authentication c. Dual key authentication d. WEP key authentication 17. Which of the following can be used with the 802.1x standard for wireless authentication? a. TACACS b. RADIUS c. Kerberos d. SSL




18. A WEP key is __________. a. Strong b. Secure c. Weak d. Safe 19. What is EAP-FAST replacing? a. PEAP b. LEAP c. WEP d. EAP-TLS 20. What IEEE group is responsible for creating WLAN security standards? a. 802.11a b. 802.11g c. 802.11ii d. 802.11i 21. Which of the following is not a correct EAP type? a. EAP-MD5 b. PEAP c. LEAP d. EAP-TTTLS 22. What security solution provides for unique changing keys that change to prevent an attacker from cracking the key? a. IPSec b. 802.1x c. Dynamic WEP d. RADIUS 23. What wireless security cipher is approved by the government for standard use? a. WEP b. RC4 c. AES d. TKIP



226


24. What IEEE standard provides the authentication framework for all IEEE 802-based networks? a. 802.11i b. 802.1x c. 802.5 d. 802.11 e. 802.11b 25. What EAP type creates an encrypted tunnel and then performs another EAP exchange inside this tunnel? a. PEAP b. EAP-TLS c. EAP-TTLS d. EAP-MD5 e. LEAP f. EAP-FAST



Chapter 13

Breaking Wireless Security This chapter looks at breaking wireless networking through the eyes of a hacker. It discusses most of the flaws that have been found relating to the many different wireless security methods and the security methods identified in the previous chapter to get a full understanding of how they can be defeated. Having an understanding of what is currently possible and what a hacker can do will help in evaluating the amount of risk surrounding any particular wireless network.

13.1 The Hacking Process This chapter defines the process taken by an attacker, a process that has been long-standing in the hacker community. With wireless networking providing a new way into networks, some of the structure of this process has changed. No longer is a network target examined and attacked. Today’s wireless war driving has shown that attackers will now find a network first, then plan an attack, and finally return to attempt a break-in. The process outlined in this chapter follows the standard hacking process, with the clear thought that a wireless network will be used as an avenue of entrance.



228


13.1.1 Information Gathering Step one in understanding wireless hackers has to do with how they would find a target as well as the information online about the target. This step is called information gathering. Most likely, a wireless hacker will find a target by war driving. In this scenario, an attacker is neither looking for something specific nor is he targeting a company directly. The attacker just happened to drive by and wants to either test out the security or is just looking for a free Internet connection. Chapter 2 discussed why war drivers do what they do and how they go about doing it. In this chapter, we expound on this by looking at the next steps that occur after detecting a wireless network. When an attacker war drives a network without planning it, this actually gives the company the upper hand. This is because the attacker’s level of knowledge about the company is very limited. This upper hand will not last long if the attacker decides to plan an attack on this network. After an attacker has decided to plan an attack, information gathering is the next step. To gather information, an attacker would use the largest amount of data ever put together, the Internet. Most people are unaware of the Internet’s ability to find information about companies or people themselves. Have you ever “googled” yourself? Try it one day with your first name and last name in quotes (e.g., “Aaron Earle”). What comes back might amaze you. An attacker will do this with a company’s name and, lo and behold, he will find a slew of information about that company. What the attacker is really looking for is IT people from the company trying to get help about a product. If someone at this company needed some help with a product, one of the steps might be to try a message board for the answer. Most likely, they will leave an e-mail address or, in some cases, the e-mail system may insert a disclosure message automatically before the e-mail leaves the system. This message most likely has the company name inside it, as well as a legal disclaimer. If an attacker is able to find this information, he can assess what types of networking products a company is using. In this case, the most relevant information would be e-mails to message boards about the company’s wireless network. The attacker can continue to gather useful information using other avenues of information gathering, such as public records. Going to the Securities & Exchange Commission’s Web site would be an example of public records. This site contains specifics on recent mergers and acquisitions, information that could serve as another avenue of attack. Also state government Web sites. Most likely, one can find information about a company and its employees as well as the names of management personnel. If two companies are linked together through a search engine, this might indicate that they are close partners. If this is true, there might also be a network connection between them. If a hacker found that the



Breaking Wireless Security

229

partner of the target has a weak network security system in place, he might use it as a backdoor into the target’s network. Another type of information-gathering technique comes from domain registration information. When companies register their domain names, they must include information about themselves. This information is publicly available and can serve as a means of social engineering. There are a number of tools to perform this type of information-gathering attack. The most common are whois, nslookup, dig, etc. A freely available tool called Sam Spade can perform all these look-ups, and more. Sam Spade is available at SamSpade.org. Web site. Once downloaded and installed, it can be run to perform any of the information-gathering techniques listed above to find domain registration information. Now take a look at Dun and Bradstreet (D&B), a company that creates, tracks, and monitors business credit. If an attacker is willing to spend around $25, he can pull a small credit report on a company. He could also spend more and get a more encompassing one that details partners, creditors, or any legal trouble the target may be experiencing. For the example here, however, the small report will work. What an attacker can do with this information is find out the business owner’s name as well as key senior members who are listed on the credit report. Once an attacker has this information, he can perform social engineering attacks on helpdesk personnel to reset passwords.

13.1.2 Enumeration Enumeration, footprinting, and scanning are all addressed in this section and defined hereafter as just enumeration. After an attacker has finished gathering information about the company he wants to attack, the next step called enumeration, takes place. In enumeration, the attacker goes looking for everything he can connect to and tries to understand what type of products there are, what company made them, and what software or firmware versions they are running. This is the step where war driving comes into play. A war driver is in the enumeration stage when he scan a wireless network. A war driver may have skipped the infor mationgathering stage altogether and go right into the enumeration stage just by war driving an area and not a selected target. A war driver may often find a network with war driving and then go back to information gathering before actually breaching the network. Looking into the enumeration stage, there is another interesting step switch or process repeat. In the enumeration stage, one may only be able to enumerate so long until one is required to move to the next step of compromise. If an attacker has exhausted his enumeration capabilities, he may have to compromise a device to gain access into another portion



230


of the network. At that time, the attacker may need to r evert to the enumeration stage to test all the new devices that he can connect to now that a certain network device has been compromised. This is very relevant in wireless network hacking. When an attacker can enumerate a wireless network, he can only go so far; after that point, he must compromise the wireless to gain access to other devices on the wired network. Once access has been gained, the enumeration process must start over on the newly discovered wired devices. There are many tools used in enumeration, most of which are some type of scanner. The different types and functions of scanner are numerous. See Chapter 15 for a discussion of a number of scanners for wireless networks. Scanning wireless networks is a small subset of what a scanner can do. The type of scanner most widely known is the port scanner, which scans for open TCP or UDP ports. After identifying these open ports, an attacker can use another scanner to finger or footprint the device by requesting and examining the information sent from these open ports. Most operating system (OS) communication across these open ports is unique to the OS manufacturer. A scanner with the ability to finger or footprint a network can examine the traffic responses from a device and understand what operating system it is by examining the traffic. This is because certain operating systems handle TCP/IP in different ways and create certain packet formats that give away the identity of the OS. Another type of scanner is one that uses ICMP and SNMP to map out networks and networking devices. This type of scanner is used to create a network map so the attacker understands the company’s network layout. This scanner is sometimes included with a port scanner and needs to run prior to a port scan to know what device to port scan. The final type of scanner discussed here is the security vulnerability scanner. This type of scanner can test a network device for a large number of known exploits and vulnerabilities. This is usually performed in a very short period of time. This is the last sub-step of enumeration; once this is done, all that is left to do is exploit the attack that the tool identified. This is the vulnerability where scanners stop short; they only identify a vulnerability and will not actually attempt to perform the next operation, called compromise. Some tools have come out that can actually execute these exploits.

13.1.3 Compromise Now the attacker attacks the wireless network or another network device to take it over; once it is taken over, it is considered compromised. If an attacker can compromise a wireless access point, he can insert himself inside the network and go after the next network target. © 2006 by Taylor & Francis Group, LLC


Breaking Wireless Security 231

The compromise section has limited tools available to perform the actual compromise. The previous section showed that some scanners could report on the vulnerabilities of a device, although the majority of them do not actually perform the vulnerability, but rather test that the vulnerability can be performed. To compromise a device, one must create or find an exploit. If one finds an exploit, it would likely have to be compiled and executed. To find these exploits, one needs to know where to look; these exploits are available in underground Web sites or IRC channels. Most of the time after an exploit has been out in public view for a considerable amount of time, a tool may appear that is point, click, and compromise. This stage of the process is where the real meat of this chapter lies. After completing the entire hacking process, one will come back to this section, looking at each of the different types of wireless networks and how they can be compromised individually. Before doing that, one needs to explore the remainder of the hacking process.

13.1.4 Expanding Privileges and Accessibility This step is very important in the hacking process. When looking at hacking from a wireless prospective, one needs to understand that to get to this stage, a compromise of the wireless network has already taken place. If a wireless network is compromised, then there is little else to do in relation to it. The attacker would most likely move on the next network target. This stage of the process has more weight in relation to servers and other network devices where access is achieved, although it is not full unrestricted access. The process of expanding privileges deals with attacking a device as a restricted user and taking advantages of flaws to elevate the privilege level to unrestricted. The next portion of this section deals with accessibility, or being able either to get back into a network or to find an easier way to control a compromised device. When one looks at getting back into a network, one starts to think about backdoors. There are many different ways to set up backdoors. Looking at wireless backdoors, an attacker could set up multiple backdoors in multiple places. In allowing easy accessibility to compromised devices, one needs to look at how this can be achieved. Using tools such as VNC and NCAT on devices can allow an attacker to receive remote shells. Once an attacker gets a remote shell, he can install or activate some type of remote management software. This software will allow for remote GUI OS manipulation. Looking at how to go about installing this is outside of the scope of this book; just be aware that it is part of the process and needs to be done in order to place some types of backdoors. © 2006 by Taylor & Francis Group, LLC


232


Now take a look at another interesting backdoor example. This has to do with dual homed wired and wireless computers, such as laptops. If an attacker can find his way into a company laptop, he might be able to find one with a network connection and a wireless connection. With almost all of today’s laptops shipping with wireless network cards, the odds of finding this is quite high and very likely. If an attacker can find this and compromises it, he can set up the wireless computer as ad hoc and set up routing into the network. Now when the attacker connects back into the network, no matter what level of wireless security is present, an attacker can circumvent it. To get access to the laptop originally, the attacker may decide to attempt to hack into it at a hotspot or steal it temporarily and return it before anyone finds out. The most likely situation would be where the attacker hacks into a network through some misconfiguration and then attacks the laptop in case the misconfiguration was detected and fixed. The next backdoor is one of the most common and, luckily for the attacker, in most cases this backdoor already exists. This backdoor is rogue access points; they are not only used by employees who want wireless, but also by attackers who stumble across them. Most of the time, these access point are deployed right out of the box with no security and default settings. This makes the network an especially easy target for an attacker. Preventing and tracking down rogue access points is addressed elsewhere in this book. The main reason that rogue access poses a threat and subsequent risk is that an attacker can use them as easy, already set-up backdoors into a corporate network. There is one other type of backdoor that is very similar to the rogue access points already discussed; expect that this one is placed by the attacker. If an attacker can gain access into the physical building, he can place a rogue access point. This is common in public buildings such as schools, hospitals, and other public-welcoming businesses. In this scenario, an attacker would purchase an inexpensive access point and find a spot to hide it. Then the attacker could mount his attack from some distance away from the physical security the building might have. With a highgain antenna connected to a laptop, an attacker could be physically located far away from any physical security force’s line of sight.

13.1.5 Cleaning up the Trails This is the last section in the hacking process; it is where one learns what the hackers do to clean up after themselves. If companies were to catch on to the existence of an attacker, it would be through log files and advanced security tools such as IDS and IPS systems. In this section, the attacker would purge logs and perform other housekeeping techniques




233

to hide their existence. This step is required in the hacking process to prevent a company from figuring out that an attacker has broken in. If this step is not performed, a large amount of evidence would remain as to how the attacker got in, where they went, and what they accessed. When we explore this section in relation to wireless, most of the same ideas and techniques apply. A wireless attacker would try to cover his tracks by changing the access point’s log file. This would have to extend to other network devices as well. The switch port would hold the attacker’s MAC although he entered through the access point. Any servers the hacker attempted to access would have log files indicating such an attempt. All of these devices would need to have their logs cleaned up. To make things even trickier is the existence of a syslog server. If a syslog server was present, all devices would send their logs to it. This means the attacker would have to attack the syslog server as well if he wanted to clean all trails of his intrusion. One of the misconceptions of the track covering section is how most hackers are found out. What some attackers tend not to realize is that if they attack a box and are unable to compromise it, their attack will be logged. That log will remain on the device for some time. Some of the ways that the log would be removed include: Someone goes through log file and manually erases it. The attacker manages to compromise the device and change the log. The log has a specific setting that allows it to be overwritten after it has taken up a specific amount of disk space. An attacker has many tools developed over the years to remove specific entries or repopulate log files with bogus entries. These tools are commonly used to cover the tracks of hackers once they have compromised the device. There are different methods as to how they work. Some tools actually change log events; they would change the event ID or some other part of the log record to make it look like it belongs. Another method involves erasing the entries themselves; some tools allow attackers to erase certain log records that deal with their access attempts. The final kind of log-cleaning tool discussed here performs this function in an obscure way. It populates the log with a large number of bogus entries to make searching for the attacker entries a long and drawn-out process.

13.2 Wireless Network Compromising Techniques This chapter section explores specific techniques used to compromise wireless networks. It details wireless security methods, looking at what



234


vulnerabilities, flaws, and exploits are out in the wild and used to compromise the different wireless security methods. With this knowledge, a proper risk assessment can be made against the many different security protocols, methods, and standards. Compromising a wireless network is only an avenue into the network. When an attacker attacks a wireless network and successfully gets access, one must understand how he got in and how to prevent his access in the future. The section will address each of the different security flaws in most of the recent wireless security methods, protocols, and standards. This information will show in detail how the exploit, problem, or attack works; how hackers use them; and what can be done to prevent them. Most of these attacks have the only secure work around of using a different security method all too together.

13.2.1 WEP WEP has undergone major scrutiny and subsequently failed all three tenets of its original design. This is observed in each of the areas of confidentiality, availability, and integrity and outlined in this section. Under confidentiality, we explor e the known plaintext attack, shar ed key authentication attack, double encryption attack, man-in-the-middle attack, and a dictionary attack. When looking at availability, we discuss a number of denial-of-service attacks that are mounted against wireless local area networks. Finally, when exploring integrity, we look at some message modification attacks.

13.2.1.1 Stream Cipher Attack The first attack against the confidentiality of wireless WEP is the stream cipher attack. This attack is based on the work and subsequent article published by Fluhrer, Mantin, and Shamir (FSM). This attack is also the basis of the WEPCrack, Airsnort, BSD-AIR Tools, and many other WEP cracking tools. The attack works in a passive manner in which no one can detect that it is even taking place. This means that trying to find out if this attack is running is nearly impossible. This works because the device running the attack does not need to transmit; instead, it only needs to receive network traffic. The attack works by taking advantage of a number of weak IV numbers that are used in the general sequence of WEP. These IVs are used as a mechanism to create a different key for each packet transmitted. There are about 9000 interesting IV numbers out of the available 16,777,216. The reasoning behind why these 9000 IV numbers are considered interesting is




235

the fact that the IV has “FF” in the middle of the IV sequence. This IV sequence is defined as three groups of two hexadecimal digits separated by a colon. An example would look similar to the following: A3:4D:33. An interesting IV would have FF in middle and look like this: 3A:FF:5E. When one looks at any IP packet, one can see that RFC 1042 requires an 802.2 SNAP header. This header is 0x88; because one knows what the plaintext is as well as the encrypted portion of it, one can get information on the first key bit. How this works is that the FF comes out to all one digits in binary and when one takes the header portion, which in the SNAP header case is inserted into the data, one gets a piece of the frame in which one knows what the first couple of bytes are before and after they are encrypted. With this knowledge, one can perform what is called a cleartext cryptanalyst attack. The topic of cleartext cryptanalyst is addressed in more detail in the next section. Once that information is known, the process to get the next key bits becomes a guessing game; however, this guessing game has a high probability of correct guessing. At first, the guessing may only yield a 5 percent chance of a correct guess, although as more key bytes are exposed, the percentage of success rapidly increases. A more detailed explanation of the mathematics of how this is achieved is available in the article entitled “Weaknesses in the Key Scheduling Algorithm of RC4.”

13.2.1.2 Known Plaintext Attack The known plaintext attack is similar to what the FSM attack uses as a starting point. This attack is possible when one knows or has access to the cleartext and the encrypted text of an information exchange. Having both the encrypted and unencrypted information allows one to perform this attack and subsequently derive a key. When the text is originally encrypted, it performs an XOR operation that mixes the key with the data. This process is reversible from encrypted text to cleartext just as it is originally applied from cleartext to encrypted text. What this means is that the same process used to encrypt can also decrypt. When the text is broken down to binary, one can perform some binary math and compute the key that was used. Looking at Table 13.1, one can see that the plaintext and key stream are broken down to binary, and binary math is performed to produce the ciphertext. If one were to place the ciphertext below the plaintext and perform the same operation, one would get the key stream. This is how the plaintext cryptanalysis attack works. To see this attack in action, there are a couple of scenarios in which it might take place. All of the scenarios of this attack fulfill the two main requirements needed: one plaintext packet and its encrypted counterpart. © 2006 by Taylor & Francis Group, LLC


236


Table 13.1

Plaintext Cryptanalysis Detail

Known Plaintext Attack Example: 1110100010101010111010

Wireless Security Handbook

Wireless Security Handbook

Wireless Security

Handbook of Research on Wireless Security

Handbook of Research on Wireless Security

Internet and Wireless Security

Wireless Operational Security

Maximum Wireless Security

Wireless Communications Security

Bulletproof Wireless Security

Wireless Network Security

Wireless network security

Cisco wireless LAN security

Mobile and Wireless Networks Security

The Wireless Data Handbook

Security in Wireless Mesh Networks

Wireless technician's handbook

Wireless Networking Handbook

Wireless Broadband Networks Handbook

Wireless Broadband Handbook

Wireless Technician's Handbook

Hacking Exposed Wireless: Wireless Security Secrets & Solutions

Hacking Exposed Wireless: Wireless Security Secrets & Solutions, 2nd edition

Hacking Exposed Wireless: Wireless Security Secrets & Solutions

Security in Wireless Mesh Networks (Wireless Networks and Mobile Communications)

Security in Wireless Mesh Networks (Wireless Networks and Mobile Communications)

Information Security Management Handbook

Computer security handbook

Java Security Handbook

Handbook of Information Security

The Home Security Handbook

Wireless Security Handbook